[InstCombine] InstCombine and InstSimplify for minimum and maximum

[tlively]

Summary

Depends on D52765

[aheejin]

lib/Analysis/InstructionSimplify.cpp

4864 ↗

(On Diff #167876)

Nit: No need to enclose within () I guess..?

lib/Analysis/ValueTracking.cpp

2904 ↗

(On Diff #167876)

I->getOperand(1)?

[tlively]

• Address comments

[tlively]

• Rebase

llvm/trunk/lib/Analysis/InstructionSimplify.cpp

if (auto *Power = dyn_cast<ConstantInt>(Op1)) {
       if (Power->isZero())
         return ConstantFP::get(Op0->getType(), 1.0);
       // powi(x, 1) -> x
       if (Power->isOne())
         return Op0;
     }
     break;
   case Intrinsic::maxnum:
-  case Intrinsic::minnum: {
+  case Intrinsic::minnum:
+  case Intrinsic::maximum:
+  case Intrinsic::minimum: {
     // If the arguments are the same, this is a no-op.
     if (Op0 == Op1) return Op0;
 
-    // If one argument is NaN or undef, return the other argument.
-    if (match(Op0, m_CombineOr(m_NaN(), m_Undef()))) return Op1;
-    if (match(Op1, m_CombineOr(m_NaN(), m_Undef()))) return Op0;
+    // If one argument is undef, return the other argument.
+    if (match(Op0, m_Undef()))
+      return Op1;
+    if (match(Op1, m_Undef()))
+      return Op0;
+
+    // If one argument is NaN, return other or NaN appropriately.
+    bool PropagateNaN = IID == Intrinsic::minimum || IID == Intrinsic::maximum;
+    if (match(Op0, m_NaN()))
+      return PropagateNaN ? Op0 : Op1;
+    if (match(Op1, m_NaN()))
+      return PropagateNaN ? Op1 : Op0;
 
     // Min/max of the same operation with common operand:
     // m(m(X, Y)), X --> m(X, Y) (4 commuted variants)
     if (auto *M0 = dyn_cast<IntrinsicInst>(Op0))
       if (M0->getIntrinsicID() == IID &&
           (M0->getOperand(0) == Op1 || M0->getOperand(1) == Op1))
         return Op0;
     if (auto *M1 = dyn_cast<IntrinsicInst>(Op1))
       if (M1->getIntrinsicID() == IID &&
           (M1->getOperand(0) == Op0 || M1->getOperand(1) == Op0))
         return Op1;
 
-    // minnum(X, -Inf) --> -Inf (and commuted variant)
-    // maxnum(X, +Inf) --> +Inf (and commuted variant)
-    bool UseNegInf = IID == Intrinsic::minnum;
+    // min(X, -Inf) --> -Inf (and commuted variant)
+    // max(X, +Inf) --> +Inf (and commuted variant)
+    bool UseNegInf = IID == Intrinsic::minnum || IID == Intrinsic::minimum;
     const APFloat *C;
     if ((match(Op0, m_APFloat(C)) && C->isInfinity() &&
          C->isNegative() == UseNegInf) ||
         (match(Op1, m_APFloat(C)) && C->isInfinity() &&
          C->isNegative() == UseNegInf))
       return ConstantFP::getInfinity(ReturnType, UseNegInf);
 
     // TODO: minnum(nnan x, inf) -> x

llvm/trunk/lib/Analysis/ValueTracking.cpp

case Instruction::Call:
     case Intrinsic::maxnum:
       return (isKnownNeverNaN(I->getOperand(0), TLI) &&
               cannotBeOrderedLessThanZeroImpl(I->getOperand(0), TLI,
                                               SignBitOnly, Depth + 1)) ||
             (isKnownNeverNaN(I->getOperand(1), TLI) &&
               cannotBeOrderedLessThanZeroImpl(I->getOperand(1), TLI,
                                               SignBitOnly, Depth + 1));
 
+    case Intrinsic::maximum:
+      return cannotBeOrderedLessThanZeroImpl(I->getOperand(0), TLI, SignBitOnly,
+                                             Depth + 1) ||
+             cannotBeOrderedLessThanZeroImpl(I->getOperand(1), TLI, SignBitOnly,
+                                             Depth + 1);
     case Intrinsic::minnum:
+    case Intrinsic::minimum:
       return cannotBeOrderedLessThanZeroImpl(I->getOperand(0), TLI, SignBitOnly,
                                              Depth + 1) &&
              cannotBeOrderedLessThanZeroImpl(I->getOperand(1), TLI, SignBitOnly,
                                              Depth + 1);
     case Intrinsic::exp:
     case Intrinsic::exp2:
     case Intrinsic::fabs:
       return true;

llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp

case Intrinsic::ssub_with_overflow: {
     if (OptimizeOverflowCheck(OCF, II->getArgOperand(0), II->getArgOperand(1),
                               *II, OperationResult, OverflowResult))
       return CreateOverflowTuple(II, OperationResult, OverflowResult);
 
     break;
   }
 
   case Intrinsic::minnum:
-  case Intrinsic::maxnum: {
+  case Intrinsic::maxnum:
+  case Intrinsic::minimum:
+  case Intrinsic::maximum: {
     Value *Arg0 = II->getArgOperand(0);
     Value *Arg1 = II->getArgOperand(1);
     // Canonicalize constants to the RHS.
     if (isa<ConstantFP>(Arg0) && !isa<ConstantFP>(Arg1)) {
       II->setArgOperand(0, Arg1);
       II->setArgOperand(1, Arg0);
       return II;
     }
 
     Value *X, *Y;
     if (match(Arg0, m_FNeg(m_Value(X))) && match(Arg1, m_FNeg(m_Value(Y))) &&
         (Arg0->hasOneUse() || Arg1->hasOneUse())) {
       // If both operands are negated, invert the call and negate the result:
-      // minnum(-X, -Y) --> -(maxnum(X, Y))
-      // maxnum(-X, -Y) --> -(minnum(X, Y))
-      Intrinsic::ID NewIID = II->getIntrinsicID() == Intrinsic::maxnum ?
-          Intrinsic::minnum : Intrinsic::maxnum;
+      // min(-X, -Y) --> -(max(X, Y))
+      // max(-X, -Y) --> -(min(X, Y))
+      Intrinsic::ID NewIID;
+      switch (II->getIntrinsicID()) {
+      case Intrinsic::maxnum:
+        NewIID = Intrinsic::minnum;
+        break;
+      case Intrinsic::minnum:
+        NewIID = Intrinsic::maxnum;
+        break;
+      case Intrinsic::maximum:
+        NewIID = Intrinsic::minimum;
+        break;
+      case Intrinsic::minimum:
+        NewIID = Intrinsic::maximum;
+        break;
+      default:
+        llvm_unreachable("unexpected intrinsic ID");
+      }
       Value *NewCall = Builder.CreateBinaryIntrinsic(NewIID, X, Y, II);
       Instruction *FNeg = BinaryOperator::CreateFNeg(NewCall);
       FNeg->copyIRFlags(II);
       return FNeg;
     }
     break;
   }
   case Intrinsic::fmuladd: {

llvm/trunk/test/Transforms/InstCombine/maximum.ll

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -S -instcombine < %s | FileCheck %s
+
+declare float @llvm.maximum.f32(float, float)
+declare <2 x float> @llvm.maximum.v2f32(<2 x float>, <2 x float>)
+declare <4 x float> @llvm.maximum.v4f32(<4 x float>, <4 x float>)
+
+declare double @llvm.maximum.f64(double, double)
+declare <2 x double> @llvm.maximum.v2f64(<2 x double>, <2 x double>)
+
+define float @constant_fold_maximum_f32() {
+; CHECK-LABEL: @constant_fold_maximum_f32(
+; CHECK-NEXT:    ret float 2.000000e+00
+;
+  %x = call float @llvm.maximum.f32(float 1.0, float 2.0)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_inv() {
+; CHECK-LABEL: @constant_fold_maximum_f32_inv(
+; CHECK-NEXT:    ret float 2.000000e+00
+;
+  %x = call float @llvm.maximum.f32(float 2.0, float 1.0)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_nan0() {
+; CHECK-LABEL: @constant_fold_maximum_f32_nan0(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %x = call float @llvm.maximum.f32(float 0x7FF8000000000000, float 2.0)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_nan1() {
+; CHECK-LABEL: @constant_fold_maximum_f32_nan1(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %x = call float @llvm.maximum.f32(float 2.0, float 0x7FF8000000000000)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_nan_nan() {
+; CHECK-LABEL: @constant_fold_maximum_f32_nan_nan(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %x = call float @llvm.maximum.f32(float 0x7FF8000000000000, float 0x7FF8000000000000)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_p0_p0() {
+; CHECK-LABEL: @constant_fold_maximum_f32_p0_p0(
+; CHECK-NEXT:    ret float 0.000000e+00
+;
+  %x = call float @llvm.maximum.f32(float 0.0, float 0.0)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_p0_n0() {
+; CHECK-LABEL: @constant_fold_maximum_f32_p0_n0(
+; CHECK-NEXT:    ret float 0.000000e+00
+;
+  %x = call float @llvm.maximum.f32(float 0.0, float -0.0)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_n0_p0() {
+; CHECK-LABEL: @constant_fold_maximum_f32_n0_p0(
+; CHECK-NEXT:    ret float 0.000000e+00
+;
+  %x = call float @llvm.maximum.f32(float -0.0, float 0.0)
+  ret float %x
+}
+
+define float @constant_fold_maximum_f32_n0_n0() {
+; CHECK-LABEL: @constant_fold_maximum_f32_n0_n0(
+; CHECK-NEXT:    ret float -0.000000e+00
+;
+  %x = call float @llvm.maximum.f32(float -0.0, float -0.0)
+  ret float %x
+}
+
+define <4 x float> @constant_fold_maximum_v4f32() {
+; CHECK-LABEL: @constant_fold_maximum_v4f32(
+; CHECK-NEXT:    ret <4 x float> <float 2.000000e+00, float 8.000000e+00, float 1.000000e+01, float 9.000000e+00>
+;
+  %x = call <4 x float> @llvm.maximum.v4f32(<4 x float> <float 1.0, float 8.0, float 3.0, float 9.0>, <4 x float> <float 2.0, float 2.0, float 10.0, float 5.0>)
+  ret <4 x float> %x
+}
+
+define double @constant_fold_maximum_f64() {
+; CHECK-LABEL: @constant_fold_maximum_f64(
+; CHECK-NEXT:    ret double 2.000000e+00
+;
+  %x = call double @llvm.maximum.f64(double 1.0, double 2.0)
+  ret double %x
+}
+
+define double @constant_fold_maximum_f64_nan0() {
+; CHECK-LABEL: @constant_fold_maximum_f64_nan0(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %x = call double @llvm.maximum.f64(double 0x7FF8000000000000, double 2.0)
+  ret double %x
+}
+
+define double @constant_fold_maximum_f64_nan1() {
+; CHECK-LABEL: @constant_fold_maximum_f64_nan1(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %x = call double @llvm.maximum.f64(double 2.0, double 0x7FF8000000000000)
+  ret double %x
+}
+
+define double @constant_fold_maximum_f64_nan_nan() {
+; CHECK-LABEL: @constant_fold_maximum_f64_nan_nan(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %x = call double @llvm.maximum.f64(double 0x7FF8000000000000, double 0x7FF8000000000000)
+  ret double %x
+}
+
+define float @canonicalize_constant_maximum_f32(float %x) {
+; CHECK-LABEL: @canonicalize_constant_maximum_f32(
+; CHECK-NEXT:    [[Y:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float 1.000000e+00)
+; CHECK-NEXT:    ret float [[Y]]
+;
+  %y = call float @llvm.maximum.f32(float 1.0, float %x)
+  ret float %y
+}
+
+define float @maximum_f32_nan_val(float %x) {
+; CHECK-LABEL: @maximum_f32_nan_val(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %y = call float @llvm.maximum.f32(float 0x7FF8000000000000, float %x)
+  ret float %y
+}
+
+define float @maximum_f32_val_nan(float %x) {
+; CHECK-LABEL: @maximum_f32_val_nan(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %y = call float @llvm.maximum.f32(float %x, float 0x7FF8000000000000)
+  ret float %y
+}
+
+define float @maximum4(float %x, float %y, float %z, float %w) {
+; CHECK-LABEL: @maximum4(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.maximum.f32(float [[Z:%.*]], float [[W:%.*]])
+; CHECK-NEXT:    [[C:%.*]] = call float @llvm.maximum.f32(float [[A]], float [[B]])
+; CHECK-NEXT:    ret float [[C]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %z, float %w)
+  %c = call float @llvm.maximum.f32(float %a, float %b)
+  ret float %c
+}
+
+; PR37404 - https://bugs.llvm.org/show_bug.cgi?id=37404
+
+define <2 x float> @neg_neg(<2 x float> %x, <2 x float> %y) {
+; CHECK-LABEL: @neg_neg(
+; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x float> @llvm.minimum.v2f32(<2 x float> [[X:%.*]], <2 x float> [[Y:%.*]])
+; CHECK-NEXT:    [[R:%.*]] = fsub <2 x float> <float -0.000000e+00, float -0.000000e+00>, [[TMP1]]
+; CHECK-NEXT:    ret <2 x float> [[R]]
+;
+  %negx = fsub <2 x float> <float -0.0, float -0.0>, %x
+  %negy = fsub <2 x float> <float -0.0, float -0.0>, %y
+  %r = call <2 x float> @llvm.maximum.v2f32(<2 x float> %negx, <2 x float> %negy)
+  ret <2 x float> %r
+}
+
+; FMF is not required, but it should be propagated from the intrinsic (not the fnegs).
+
+define float @neg_neg_vec_fmf(float %x, float %y) {
+; CHECK-LABEL: @neg_neg_vec_fmf(
+; CHECK-NEXT:    [[TMP1:%.*]] = call fast float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[R:%.*]] = fsub fast float -0.000000e+00, [[TMP1]]
+; CHECK-NEXT:    ret float [[R]]
+;
+  %negx = fsub arcp float -0.0, %x
+  %negy = fsub afn float -0.0, %y
+  %r = call fast float @llvm.maximum.f32(float %negx, float %negy)
+  ret float %r
+}
+
+; 1 extra use of an intermediate value should still allow the fold,
+; but 2 would require more instructions than we started with.
+
+declare void @use(float)
+define float @neg_neg_extra_use_x(float %x, float %y) {
+; CHECK-LABEL: @neg_neg_extra_use_x(
+; CHECK-NEXT:    [[NEGX:%.*]] = fsub float -0.000000e+00, [[X:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.minimum.f32(float [[X]], float [[Y:%.*]])
+; CHECK-NEXT:    [[R:%.*]] = fsub float -0.000000e+00, [[TMP1]]
+; CHECK-NEXT:    call void @use(float [[NEGX]])
+; CHECK-NEXT:    ret float [[R]]
+;
+  %negx = fsub float -0.0, %x
+  %negy = fsub float -0.0, %y
+  %r = call float @llvm.maximum.f32(float %negx, float %negy)
+  call void @use(float %negx)
+  ret float %r
+}
+
+define float @neg_neg_extra_use_y(float %x, float %y) {
+; CHECK-LABEL: @neg_neg_extra_use_y(
+; CHECK-NEXT:    [[NEGY:%.*]] = fsub float -0.000000e+00, [[Y:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y]])
+; CHECK-NEXT:    [[R:%.*]] = fsub float -0.000000e+00, [[TMP1]]
+; CHECK-NEXT:    call void @use(float [[NEGY]])
+; CHECK-NEXT:    ret float [[R]]
+;
+  %negx = fsub float -0.0, %x
+  %negy = fsub float -0.0, %y
+  %r = call float @llvm.maximum.f32(float %negx, float %negy)
+  call void @use(float %negy)
+  ret float %r
+}
+
+define float @neg_neg_extra_use_x_and_y(float %x, float %y) {
+; CHECK-LABEL: @neg_neg_extra_use_x_and_y(
+; CHECK-NEXT:    [[NEGX:%.*]] = fsub float -0.000000e+00, [[X:%.*]]
+; CHECK-NEXT:    [[NEGY:%.*]] = fsub float -0.000000e+00, [[Y:%.*]]
+; CHECK-NEXT:    [[R:%.*]] = call float @llvm.maximum.f32(float [[NEGX]], float [[NEGY]])
+; CHECK-NEXT:    call void @use(float [[NEGX]])
+; CHECK-NEXT:    call void @use(float [[NEGY]])
+; CHECK-NEXT:    ret float [[R]]
+;
+  %negx = fsub float -0.0, %x
+  %negy = fsub float -0.0, %y
+  %r = call float @llvm.maximum.f32(float %negx, float %negy)
+  call void @use(float %negx)
+  call void @use(float %negy)
+  ret float %r
+}

llvm/trunk/test/Transforms/InstCombine/minimum.ll

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -S -instcombine < %s | FileCheck %s
+
+declare float @llvm.minimum.f32(float, float)
+declare float @llvm.minimum.v2f32(<2 x float>, <2 x float>)
+declare <4 x float> @llvm.minimum.v4f32(<4 x float>, <4 x float>)
+
+declare double @llvm.minimum.f64(double, double)
+declare <2 x double> @llvm.minimum.v2f64(<2 x double>, <2 x double>)
+
+declare float @llvm.maximum.f32(float, float)
+
+define float @constant_fold_minimum_f32() {
+; CHECK-LABEL: @constant_fold_minimum_f32(
+; CHECK-NEXT:    ret float 1.000000e+00
+;
+  %x = call float @llvm.minimum.f32(float 1.0, float 2.0)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_inv() {
+; CHECK-LABEL: @constant_fold_minimum_f32_inv(
+; CHECK-NEXT:    ret float 1.000000e+00
+;
+  %x = call float @llvm.minimum.f32(float 2.0, float 1.0)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_nan0() {
+; CHECK-LABEL: @constant_fold_minimum_f32_nan0(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %x = call float @llvm.minimum.f32(float 0x7FF8000000000000, float 2.0)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_nan1() {
+; CHECK-LABEL: @constant_fold_minimum_f32_nan1(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %x = call float @llvm.minimum.f32(float 2.0, float 0x7FF8000000000000)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_nan_nan() {
+; CHECK-LABEL: @constant_fold_minimum_f32_nan_nan(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %x = call float @llvm.minimum.f32(float 0x7FF8000000000000, float 0x7FF8000000000000)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_p0_p0() {
+; CHECK-LABEL: @constant_fold_minimum_f32_p0_p0(
+; CHECK-NEXT:    ret float 0.000000e+00
+;
+  %x = call float @llvm.minimum.f32(float 0.0, float 0.0)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_p0_n0() {
+; CHECK-LABEL: @constant_fold_minimum_f32_p0_n0(
+; CHECK-NEXT:    ret float -0.000000e+00
+;
+  %x = call float @llvm.minimum.f32(float 0.0, float -0.0)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_n0_p0() {
+; CHECK-LABEL: @constant_fold_minimum_f32_n0_p0(
+; CHECK-NEXT:    ret float -0.000000e+00
+;
+  %x = call float @llvm.minimum.f32(float -0.0, float 0.0)
+  ret float %x
+}
+
+define float @constant_fold_minimum_f32_n0_n0() {
+; CHECK-LABEL: @constant_fold_minimum_f32_n0_n0(
+; CHECK-NEXT:    ret float -0.000000e+00
+;
+  %x = call float @llvm.minimum.f32(float -0.0, float -0.0)
+  ret float %x
+}
+
+define <4 x float> @constant_fold_minimum_v4f32() {
+; CHECK-LABEL: @constant_fold_minimum_v4f32(
+; CHECK-NEXT:    ret <4 x float> <float 1.000000e+00, float 2.000000e+00, float 3.000000e+00, float 5.000000e+00>
+;
+  %x = call <4 x float> @llvm.minimum.v4f32(<4 x float> <float 1.0, float 8.0, float 3.0, float 9.0>, <4 x float> <float 2.0, float 2.0, float 10.0, float 5.0>)
+  ret <4 x float> %x
+}
+
+define double @constant_fold_minimum_f64() {
+; CHECK-LABEL: @constant_fold_minimum_f64(
+; CHECK-NEXT:    ret double 1.000000e+00
+;
+  %x = call double @llvm.minimum.f64(double 1.0, double 2.0)
+  ret double %x
+}
+
+define double @constant_fold_minimum_f64_nan0() {
+; CHECK-LABEL: @constant_fold_minimum_f64_nan0(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %x = call double @llvm.minimum.f64(double 0x7FF8000000000000, double 2.0)
+  ret double %x
+}
+
+define double @constant_fold_minimum_f64_nan1() {
+; CHECK-LABEL: @constant_fold_minimum_f64_nan1(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %x = call double @llvm.minimum.f64(double 2.0, double 0x7FF8000000000000)
+  ret double %x
+}
+
+define double @constant_fold_minimum_f64_nan_nan() {
+; CHECK-LABEL: @constant_fold_minimum_f64_nan_nan(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %x = call double @llvm.minimum.f64(double 0x7FF8000000000000, double 0x7FF8000000000000)
+  ret double %x
+}
+
+define float @canonicalize_constant_minimum_f32(float %x) {
+; CHECK-LABEL: @canonicalize_constant_minimum_f32(
+; CHECK-NEXT:    [[Y:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float 1.000000e+00)
+; CHECK-NEXT:    ret float [[Y]]
+;
+  %y = call float @llvm.minimum.f32(float 1.0, float %x)
+  ret float %y
+}
+
+define float @minimum_f32_nan_val(float %x) {
+; CHECK-LABEL: @minimum_f32_nan_val(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %y = call float @llvm.minimum.f32(float 0x7FF8000000000000, float %x)
+  ret float %y
+}
+
+define float @minimum_f32_val_nan(float %x) {
+; CHECK-LABEL: @minimum_f32_val_nan(
+; CHECK-NEXT:    ret float 0x7FF8000000000000
+;
+  %y = call float @llvm.minimum.f32(float %x, float 0x7FF8000000000000)
+  ret float %y
+}
+
+define float @minimum4(float %x, float %y, float %z, float %w) {
+; CHECK-LABEL: @minimum4(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.minimum.f32(float [[Z:%.*]], float [[W:%.*]])
+; CHECK-NEXT:    [[C:%.*]] = call float @llvm.minimum.f32(float [[A]], float [[B]])
+; CHECK-NEXT:    ret float [[C]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %z, float %w)
+  %c = call float @llvm.minimum.f32(float %a, float %b)
+  ret float %c
+}
+
+define float @minimum_x_maximum_x_y(float %x, float %y) {
+; CHECK-LABEL: @minimum_x_maximum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.minimum.f32(float [[X]], float [[A]])
+; CHECK-NEXT:    ret float [[B]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %x, float %a)
+  ret float %b
+}
+
+define float @maximum_x_minimum_x_y(float %x, float %y) {
+; CHECK-LABEL: @maximum_x_minimum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.maximum.f32(float [[X]], float [[A]])
+; CHECK-NEXT:    ret float [[B]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %x, float %a)
+  ret float %b
+}
+
+; PR37405 - https://bugs.llvm.org/show_bug.cgi?id=37405
+
+define double @neg_neg(double %x, double %y) {
+; CHECK-LABEL: @neg_neg(
+; CHECK-NEXT:    [[TMP1:%.*]] = call double @llvm.maximum.f64(double [[X:%.*]], double [[Y:%.*]])
+; CHECK-NEXT:    [[R:%.*]] = fsub double -0.000000e+00, [[TMP1]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %negx = fsub double -0.0, %x
+  %negy = fsub double -0.0, %y
+  %r = call double @llvm.minimum.f64(double %negx, double %negy)
+  ret double %r
+}
+
+; FMF is not required, but it should be propagated from the intrinsic (not the fnegs).
+; Also, make sure this works with vectors.
+
+define <2 x double> @neg_neg_vec_fmf(<2 x double> %x, <2 x double> %y) {
+; CHECK-LABEL: @neg_neg_vec_fmf(
+; CHECK-NEXT:    [[TMP1:%.*]] = call nnan ninf <2 x double> @llvm.maximum.v2f64(<2 x double> [[X:%.*]], <2 x double> [[Y:%.*]])
+; CHECK-NEXT:    [[R:%.*]] = fsub nnan ninf <2 x double> <double -0.000000e+00, double -0.000000e+00>, [[TMP1]]
+; CHECK-NEXT:    ret <2 x double> [[R]]
+;
+  %negx = fsub reassoc <2 x double> <double -0.0, double -0.0>, %x
+  %negy = fsub fast <2 x double> <double -0.0, double -0.0>, %y
+  %r = call nnan ninf <2 x double> @llvm.minimum.v2f64(<2 x double> %negx, <2 x double> %negy)
+  ret <2 x double> %r
+}
+
+; 1 extra use of an intermediate value should still allow the fold,
+; but 2 would require more instructions than we started with.
+
+declare void @use(double)
+define double @neg_neg_extra_use_x(double %x, double %y) {
+; CHECK-LABEL: @neg_neg_extra_use_x(
+; CHECK-NEXT:    [[NEGX:%.*]] = fsub double -0.000000e+00, [[X:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call double @llvm.maximum.f64(double [[X]], double [[Y:%.*]])
+; CHECK-NEXT:    [[R:%.*]] = fsub double -0.000000e+00, [[TMP1]]
+; CHECK-NEXT:    call void @use(double [[NEGX]])
+; CHECK-NEXT:    ret double [[R]]
+;
+  %negx = fsub double -0.0, %x
+  %negy = fsub double -0.0, %y
+  %r = call double @llvm.minimum.f64(double %negx, double %negy)
+  call void @use(double %negx)
+  ret double %r
+}
+
+define double @neg_neg_extra_use_y(double %x, double %y) {
+; CHECK-LABEL: @neg_neg_extra_use_y(
+; CHECK-NEXT:    [[NEGY:%.*]] = fsub double -0.000000e+00, [[Y:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call double @llvm.maximum.f64(double [[X:%.*]], double [[Y]])
+; CHECK-NEXT:    [[R:%.*]] = fsub double -0.000000e+00, [[TMP1]]
+; CHECK-NEXT:    call void @use(double [[NEGY]])
+; CHECK-NEXT:    ret double [[R]]
+;
+  %negx = fsub double -0.0, %x
+  %negy = fsub double -0.0, %y
+  %r = call double @llvm.minimum.f64(double %negx, double %negy)
+  call void @use(double %negy)
+  ret double %r
+}
+
+define double @neg_neg_extra_use_x_and_y(double %x, double %y) {
+; CHECK-LABEL: @neg_neg_extra_use_x_and_y(
+; CHECK-NEXT:    [[NEGX:%.*]] = fsub double -0.000000e+00, [[X:%.*]]
+; CHECK-NEXT:    [[NEGY:%.*]] = fsub double -0.000000e+00, [[Y:%.*]]
+; CHECK-NEXT:    [[R:%.*]] = call double @llvm.minimum.f64(double [[NEGX]], double [[NEGY]])
+; CHECK-NEXT:    call void @use(double [[NEGX]])
+; CHECK-NEXT:    call void @use(double [[NEGY]])
+; CHECK-NEXT:    ret double [[R]]
+;
+  %negx = fsub double -0.0, %x
+  %negy = fsub double -0.0, %y
+  %r = call double @llvm.minimum.f64(double %negx, double %negy)
+  call void @use(double %negx)
+  call void @use(double %negy)
+  ret double %r
+}

llvm/trunk/test/Transforms/InstSimplify/floating-point-arithmetic.ll

 ; CHECK-LABEL: @maxnum_neginf(
 ; CHECK-NEXT:    [[VAL:%.*]] = call float @llvm.maxnum.f32(float 0xFFF0000000000000, float [[X:%.*]])
 ; CHECK-NEXT:    ret float [[VAL]]
 ;
   %val = call float @llvm.maxnum.f32(float 0xFFF0000000000000, float %x)
   ret float %val
 }
 
+declare float @llvm.minimum.f32(float, float)
+declare float @llvm.maximum.f32(float, float)
+declare double @llvm.minimum.f64(double, double)
+declare double @llvm.maximum.f64(double, double)
+declare <2 x double> @llvm.minimum.v2f64(<2 x double>, <2 x double>)
+declare <2 x double> @llvm.maximum.v2f64(<2 x double>, <2 x double>)
+
+; From the LangRef for minimum/maximum:
+; "If either operand is a NaN, returns NaN."
+
+define double @maximum_nan_op0(double %x) {
+; CHECK-LABEL: @maximum_nan_op0(
+; CHECK-NEXT:    ret double 0x7FF8000000000000
+;
+  %r = call double @llvm.maximum.f64(double 0x7ff8000000000000, double %x)
+  ret double %r
+}
+
+define double @maximum_nan_op1(double %x) {
+; CHECK-LABEL: @maximum_nan_op1(
+; CHECK-NEXT:    ret double 0x7FF800000000DEAD
+;
+  %r = call double @llvm.maximum.f64(double %x, double 0x7ff800000000dead)
+  ret double %r
+}
+
+define double @minimum_nan_op0(double %x) {
+; CHECK-LABEL: @minimum_nan_op0(
+; CHECK-NEXT:    ret double 0x7FF8000DEAD00000
+;
+  %r = call double @llvm.minimum.f64(double 0x7ff8000dead00000, double %x)
+  ret double %r
+}
+
+define double @minimum_nan_op1(double %x) {
+; CHECK-LABEL: @minimum_nan_op1(
+; CHECK-NEXT:    ret double 0x7FF800DEAD00DEAD
+;
+  %r = call double @llvm.minimum.f64(double %x, double 0x7ff800dead00dead)
+  ret double %r
+}
+
+define <2 x double> @maximum_nan_op0_vec(<2 x double> %x) {
+; CHECK-LABEL: @maximum_nan_op0_vec(
+; CHECK-NEXT:    ret <2 x double> <double 0x7FF8000000000000, double undef>
+;
+  %r = call <2 x double> @llvm.maximum.v2f64(<2 x double> <double 0x7ff8000000000000, double undef>, <2 x double> %x)
+  ret <2 x double> %r
+}
+
+define <2 x double> @maximum_nan_op1_vec(<2 x double> %x) {
+; CHECK-LABEL: @maximum_nan_op1_vec(
+; CHECK-NEXT:    ret <2 x double> <double 0x7FF800000000DEAD, double 0x7FF8FFFFFFFFFFFF>
+;
+  %r = call <2 x double> @llvm.maximum.v2f64(<2 x double> %x, <2 x double> <double 0x7ff800000000dead, double 0x7ff8ffffffffffff>)
+  ret <2 x double> %r
+}
+
+define <2 x double> @minimum_nan_op0_vec(<2 x double> %x) {
+; CHECK-LABEL: @minimum_nan_op0_vec(
+; CHECK-NEXT:    ret <2 x double> <double undef, double 0x7FF8000DEAD00000>
+;
+  %r = call <2 x double> @llvm.minimum.v2f64(<2 x double> <double undef, double 0x7ff8000dead00000>, <2 x double> %x)
+  ret <2 x double> %r
+}
+
+define <2 x double> @minimum_nan_op1_vec(<2 x double> %x) {
+; CHECK-LABEL: @minimum_nan_op1_vec(
+; CHECK-NEXT:    ret <2 x double> <double 0x7FF800DEAD00DEAD, double 0x7FF800DEAD00DEAD>
+;
+  %r = call <2 x double> @llvm.minimum.v2f64(<2 x double> %x, <2 x double> <double 0x7ff800dead00dead, double 0x7ff800dead00dead>)
+  ret <2 x double> %r
+}
+
+define float @maximum_undef_op1(float %x) {
+; CHECK-LABEL: @maximum_undef_op1(
+; CHECK-NEXT:    ret float [[X:%.*]]
+;
+  %val = call float @llvm.maximum.f32(float %x, float undef)
+  ret float %val
+}
+
+define float @maximum_undef_op0(float %x) {
+; CHECK-LABEL: @maximum_undef_op0(
+; CHECK-NEXT:    ret float [[X:%.*]]
+;
+  %val = call float @llvm.maximum.f32(float undef, float %x)
+  ret float %val
+}
+
+define float @minimum_undef_op1(float %x) {
+; CHECK-LABEL: @minimum_undef_op1(
+; CHECK-NEXT:    ret float [[X:%.*]]
+;
+  %val = call float @llvm.minimum.f32(float %x, float undef)
+  ret float %val
+}
+
+define float @minimum_undef_op0(float %x) {
+; CHECK-LABEL: @minimum_undef_op0(
+; CHECK-NEXT:    ret float [[X:%.*]]
+;
+  %val = call float @llvm.minimum.f32(float undef, float %x)
+  ret float %val
+}
+
+define float @minimum_undef_undef(float %x) {
+; CHECK-LABEL: @minimum_undef_undef(
+; CHECK-NEXT:    ret float undef
+;
+  %val = call float @llvm.minimum.f32(float undef, float undef)
+  ret float %val
+}
+
+define float @maximum_undef_undef(float %x) {
+; CHECK-LABEL: @maximum_undef_undef(
+; CHECK-NEXT:    ret float undef
+;
+  %val = call float @llvm.maximum.f32(float undef, float undef)
+  ret float %val
+}
+
+define float @minimum_same_args(float %x) {
+; CHECK-LABEL: @minimum_same_args(
+; CHECK-NEXT:    ret float [[X:%.*]]
+;
+  %y = call float @llvm.minimum.f32(float %x, float %x)
+  ret float %y
+}
+
+define float @maximum_same_args(float %x) {
+; CHECK-LABEL: @maximum_same_args(
+; CHECK-NEXT:    ret float [[X:%.*]]
+;
+  %y = call float @llvm.maximum.f32(float %x, float %x)
+  ret float %y
+}
+
+define float @minimum_x_minimum_x_y(float %x, float %y) {
+; CHECK-LABEL: @minimum_x_minimum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %x, float %a)
+  ret float %b
+}
+
+define float @minimum_y_minimum_x_y(float %x, float %y) {
+; CHECK-LABEL: @minimum_y_minimum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %y, float %a)
+  ret float %b
+}
+
+define float @minimum_x_y_minimum_x(float %x, float %y) {
+; CHECK-LABEL: @minimum_x_y_minimum_x(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %a, float %x)
+  ret float %b
+}
+
+define float @minimum_x_y_minimum_y(float %x, float %y) {
+; CHECK-LABEL: @minimum_x_y_minimum_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %a, float %y)
+  ret float %b
+}
+
+; negative test
+
+define float @minimum_z_minimum_x_y(float %x, float %y, float %z) {
+; CHECK-LABEL: @minimum_z_minimum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.minimum.f32(float [[Z:%.*]], float [[A]])
+; CHECK-NEXT:    ret float [[B]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %z, float %a)
+  ret float %b
+}
+
+; negative test
+
+define float @minimum_x_y_minimum_z(float %x, float %y, float %z) {
+; CHECK-LABEL: @minimum_x_y_minimum_z(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.minimum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.minimum.f32(float [[A]], float [[Z:%.*]])
+; CHECK-NEXT:    ret float [[B]]
+;
+  %a = call float @llvm.minimum.f32(float %x, float %y)
+  %b = call float @llvm.minimum.f32(float %a, float %z)
+  ret float %b
+}
+
+; minimum(X, -INF) --> -INF
+
+define float @minimum_neginf(float %x) {
+; CHECK-LABEL: @minimum_neginf(
+; CHECK-NEXT:    ret float 0xFFF0000000000000
+;
+  %val = call float @llvm.minimum.f32(float %x, float 0xFFF0000000000000)
+  ret float %val
+}
+
+define <2 x double> @minimum_neginf_commute_vec(<2 x double> %x) {
+; CHECK-LABEL: @minimum_neginf_commute_vec(
+; CHECK-NEXT:    ret <2 x double> <double 0xFFF0000000000000, double 0xFFF0000000000000>
+;
+  %r = call <2 x double> @llvm.minimum.v2f64(<2 x double> <double 0xFFF0000000000000, double 0xFFF0000000000000>, <2 x double> %x)
+  ret <2 x double> %r
+}
+
+; negative test
+
+define float @minimum_inf(float %x) {
+; CHECK-LABEL: @minimum_inf(
+; CHECK-NEXT:    [[VAL:%.*]] = call float @llvm.minimum.f32(float 0x7FF0000000000000, float [[X:%.*]])
+; CHECK-NEXT:    ret float [[VAL]]
+;
+  %val = call float @llvm.minimum.f32(float 0x7FF0000000000000, float %x)
+  ret float %val
+}
+define float @maximum_x_maximum_x_y(float %x, float %y) {
+; CHECK-LABEL: @maximum_x_maximum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %x, float %a)
+  ret float %b
+}
+
+define float @maximum_y_maximum_x_y(float %x, float %y) {
+; CHECK-LABEL: @maximum_y_maximum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %y, float %a)
+  ret float %b
+}
+
+define float @maximum_x_y_maximum_x(float %x, float %y) {
+; CHECK-LABEL: @maximum_x_y_maximum_x(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %a, float %x)
+  ret float %b
+}
+
+define float @maximum_x_y_maximum_y(float %x, float %y) {
+; CHECK-LABEL: @maximum_x_y_maximum_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    ret float [[A]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %a, float %y)
+  ret float %b
+}
+
+; negative test
+
+define float @maximum_z_maximum_x_y(float %x, float %y, float %z) {
+; CHECK-LABEL: @maximum_z_maximum_x_y(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.maximum.f32(float [[Z:%.*]], float [[A]])
+; CHECK-NEXT:    ret float [[B]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %z, float %a)
+  ret float %b
+}
+
+; negative test
+
+define float @maximum_x_y_maximum_z(float %x, float %y, float %z) {
+; CHECK-LABEL: @maximum_x_y_maximum_z(
+; CHECK-NEXT:    [[A:%.*]] = call float @llvm.maximum.f32(float [[X:%.*]], float [[Y:%.*]])
+; CHECK-NEXT:    [[B:%.*]] = call float @llvm.maximum.f32(float [[A]], float [[Z:%.*]])
+; CHECK-NEXT:    ret float [[B]]
+;
+  %a = call float @llvm.maximum.f32(float %x, float %y)
+  %b = call float @llvm.maximum.f32(float %a, float %z)
+  ret float %b
+}
+
+; maximum(X, INF) --> INF
+
+define <2 x double> @maximum_inf(<2 x double> %x) {
+; CHECK-LABEL: @maximum_inf(
+; CHECK-NEXT:    ret <2 x double> <double 0x7FF0000000000000, double 0x7FF0000000000000>
+;
+  %val = call <2 x double> @llvm.maximum.v2f64(<2 x double> %x, <2 x double><double 0x7FF0000000000000, double 0x7FF0000000000000>)
+  ret <2 x double> %val
+}
+
+define float @maximum_inf_commute(float %x) {
+; CHECK-LABEL: @maximum_inf_commute(
+; CHECK-NEXT:    ret float 0x7FF0000000000000
+;
+  %val = call float @llvm.maximum.f32(float 0x7FF0000000000000, float %x)
+  ret float %val
+}
+
 ; Y - (Y - X) --> X
 
 define float @fsub_fsub_common_op(float %x, float %y) {
 ; CHECK-LABEL: @fsub_fsub_common_op(
 ; CHECK-NEXT:    ret float [[X:%.*]]
 ;
   %s = fsub float %y, %x
   %r = fsub reassoc nsz float %y, %s

 ; CHECK-NEXT:    [[S:%.*]] = fsub float [[Y:%.*]], [[X:%.*]]
 ; CHECK-NEXT:    [[R:%.*]] = fadd reassoc nsz float [[S]], [[Y]]
 ; CHECK-NEXT:    ret float [[R]]
 ;
   %s = fsub float %y, %x
   %r = fadd reassoc nsz float %s, %y
   ret float %r
 }
-

llvm/trunk/test/Transforms/InstSimplify/floating-point-compare.ll

 declare double @llvm.fabs.f64(double)
 declare <2 x float> @llvm.fabs.v2f32(<2 x float>)
 declare <2 x double> @llvm.fabs.v2f64(<2 x double>)
 declare float @llvm.sqrt.f32(float)
 declare double @llvm.powi.f64(double,i32)
 declare float @llvm.exp.f32(float)
 declare float @llvm.minnum.f32(float, float)
 declare float @llvm.maxnum.f32(float, float)
+declare float @llvm.maximum.f32(float, float)
 declare double @llvm.exp2.f64(double)
 declare float @llvm.fma.f32(float,float,float)
 
 declare void @expect_equal(i1,i1)
 
 define i1 @orderedLessZeroTree(float,float,float,float) {
 ; CHECK-LABEL: @orderedLessZeroTree(
 ; CHECK-NEXT:    ret i1 true

 ; CHECK-NEXT:    ret i1 [[UGE]]
 ;
   %a = call float @llvm.exp.f32(float %0)
   %b = call float @llvm.maxnum.f32(float %a, float %1)
   %uge = fcmp uge float %b, 0.000000e+00
   ret i1 %uge
 }
 
+; But using maximum, we can simplify, since the NaN would be propagated
+
+define i1 @orderedLessZeroMaximum(float, float) {
+; CHECK-LABEL: @orderedLessZeroMaximum(
+; CHECK-NEXT:    ret i1 true
+;
+  %a = call float @llvm.exp.f32(float %0)
+  %b = call float @llvm.maximum.f32(float %a, float %1)
+  %uge = fcmp uge float %b, 0.000000e+00
+  ret i1 %uge
+}
+
 define i1 @known_positive_olt_with_negative_constant(double %a) {
 ; CHECK-LABEL: @known_positive_olt_with_negative_constant(
 ; CHECK-NEXT:    ret i1 false
 ;
   %call = call double @llvm.fabs.f64(double %a)
   %cmp = fcmp olt double %call, -1.0
   ret i1 %cmp
 }

 
 define <2 x i1> @unorderedCompareWithNaNVector_undef_elt(<2 x double> %A) {
 ; CHECK-LABEL: @unorderedCompareWithNaNVector_undef_elt(
 ; CHECK-NEXT:    ret <2 x i1> <i1 true, i1 true>
 ;
   %cmp = fcmp ult <2 x double> %A, <double undef, double 0xFFFFFFFFFFFFFFFF>
   ret <2 x i1> %cmp
 }
-

llvm/trunk/test/Transforms/LICM/hoist-round.ll

 ; RUN: opt -S -licm < %s | FileCheck %s
 ; RUN: opt -aa-pipeline=basic-aa -passes='require<aa>,require<targetir>,require<scalar-evolution>,require<opt-remark-emit>,loop(licm)' -S %s | FileCheck %s
 
 target datalayout = "E-m:e-p:32:32-i8:8:8-i16:16:16-i64:32:32-f64:32:32-v64:32:32-v128:32:32-a0:0:32-n32"
 
 ; This test verifies that ceil, floor, nearbyint, trunc, rint, round,
-; copysign, minnum, maxnum and fabs intrinsics are considered safe
-; to speculate.
+; copysign, minnum, maxnum, minimum, maximum, and fabs intrinsics are
+; considered safe to speculate.
 
 ; CHECK-LABEL: @test
 ; CHECK: call float @llvm.ceil.f32
 ; CHECK: call float @llvm.floor.f32
 ; CHECK: call float @llvm.nearbyint.f32
 ; CHECK: call float @llvm.rint.f32
 ; CHECK: call float @llvm.round.f32
 ; CHECK: call float @llvm.trunc.f32
for.body:
   %tmp.3 = call float @llvm.nearbyint.f32(float %tmp.2)
   %tmp.4 = call float @llvm.rint.f32(float %tmp.3)
   %tmp.5 = call float @llvm.round.f32(float %tmp.4)
   %tmp.6 = call float @llvm.trunc.f32(float %tmp.5)
   %tmp.7 = call float @llvm.fabs.f32(float %tmp.6)
   %tmp.8 = call float @llvm.copysign.f32(float %tmp.7, float %arg2)
   %tmp.9 = call float @llvm.minnum.f32(float %tmp.8, float %arg2)
   %tmp.10 = call float @llvm.maxnum.f32(float %tmp.9, float %arg2)
-  %tmp.11 = call float @llvm.powi.f32(float %tmp.10, i32 4)
-  call void @consume(float %tmp.11)
+  %tmp.11 = call float @llvm.minimum.f32(float %tmp.10, float %arg2)
+  %tmp.12 = call float @llvm.maximum.f32(float %tmp.11, float %arg2)
+  %tmp.13 = call float @llvm.powi.f32(float %tmp.12, i32 4)
+  call void @consume(float %tmp.13)
   %IND.new = add i32 %IND, 1
   br label %for.head
 
 exit:
   ret void
 }
 
 declare void @consume(float)
 
 declare float @llvm.ceil.f32(float)
 declare float @llvm.floor.f32(float)
 declare float @llvm.nearbyint.f32(float)
 declare float @llvm.rint.f32(float)
 declare float @llvm.round.f32(float)
 declare float @llvm.trunc.f32(float)
 declare float @llvm.fabs.f32(float)
 declare float @llvm.copysign.f32(float, float)
 declare float @llvm.minnum.f32(float, float)
 declare float @llvm.maxnum.f32(float, float)
+declare float @llvm.minimum.f32(float, float)
+declare float @llvm.maximum.f32(float, float)
 declare float @llvm.powi.f32(float, i32)

llvm/trunk/test/Transforms/SimplifyCFG/speculate-math.ll

 ; RUN: opt -S -simplifycfg < %s | FileCheck %s --check-prefix=EXPENSIVE --check-prefix=ALL
 ; RUN: opt -S -simplifycfg -speculate-one-expensive-inst=false < %s | FileCheck %s --check-prefix=CHEAP --check-prefix=ALL
 
 declare float @llvm.sqrt.f32(float) nounwind readonly
 declare float @llvm.fma.f32(float, float, float) nounwind readonly
 declare float @llvm.fmuladd.f32(float, float, float) nounwind readonly
 declare float @llvm.fabs.f32(float) nounwind readonly
 declare float @llvm.minnum.f32(float, float) nounwind readonly
 declare float @llvm.maxnum.f32(float, float) nounwind readonly
+declare float @llvm.minimum.f32(float, float) nounwind readonly
+declare float @llvm.maximum.f32(float, float) nounwind readonly
 
 ; ALL-LABEL: @fdiv_test(
 ; EXPENSIVE: select i1 %cmp, double %div, double 0.0
 ; CHEAP-NOT: select
 
 define double @fdiv_test(double %a, double %b) {
 entry:
   %cmp = fcmp ogt double %a, 0.0
cond.else.i:                                      ; preds = %entry
   %0 = tail call float @llvm.maxnum.f32(float %a, float %b) nounwind readnone
   br label %test_maxnum.exit
 
 test_maxnum.exit:                                   ; preds = %cond.else.i, %entry
   %cond.i = phi float [ %0, %cond.else.i ], [ 0x7FF8000000000000, %entry ]
   store float %cond.i, float addrspace(1)* %out, align 4
   ret void
 }
+
+; ALL-LABEL: @minimum_test(
+; ALL: select
+define void @minimum_test(float addrspace(1)* noalias nocapture %out, float %a, float %b) nounwind {
+entry:
+  %cmp.i = fcmp olt float %a, 0.000000e+00
+  br i1 %cmp.i, label %test_minimum.exit, label %cond.else.i
+
+cond.else.i:                                      ; preds = %entry
+  %0 = tail call float @llvm.minimum.f32(float %a, float %b) nounwind readnone
+  br label %test_minimum.exit
+
+test_minimum.exit:                                   ; preds = %cond.else.i, %entry
+  %cond.i = phi float [ %0, %cond.else.i ], [ 0x7FF8000000000000, %entry ]
+  store float %cond.i, float addrspace(1)* %out, align 4
+  ret void
+}
+
+; ALL-LABEL: @maximum_test(
+; ALL: select
+define void @maximum_test(float addrspace(1)* noalias nocapture %out, float %a, float %b) nounwind {
+entry:
+  %cmp.i = fcmp olt float %a, 0.000000e+00
+  br i1 %cmp.i, label %test_maximum.exit, label %cond.else.i
+
+cond.else.i:                                      ; preds = %entry
+  %0 = tail call float @llvm.maximum.f32(float %a, float %b) nounwind readnone
+  br label %test_maximum.exit
+
+test_maximum.exit:                                   ; preds = %cond.else.i, %entry
+  %cond.i = phi float [ %0, %cond.else.i ], [ 0x7FF8000000000000, %entry ]
+  store float %cond.i, float addrspace(1)* %out, align 4
+  ret void
+}