Propagate Undef in llvm.cos Intrinsic

[anna]

Summary

the llvm cos intrinsic currently does not propagate undef's. This change transforms cos(undef) to null value or 0.

There are 2 test cases added as well.

[sanjoy]

@majnemer - what do you think of this?

lib/Analysis/ConstantFolding.cpp

1446

Since you're not using Op, using isa instead of dyn_cast is better.

[anna]

lib/Analysis/ConstantFolding.cpp

1446

Updated the code to use isa.

[sanjoy]

LGTM pending a minor comment inline. Let me know once you're ready for me to commit this for you.

test/Transforms/InstCombine/cos_intrinsic.ll

9

Can you please change this directly CHECK for the result of the fold?

[anna]

updated based on Sanjoy's comment to place the check in the first test after the cos(undef) itself

lib/Analysis/ConstantFolding.cpp

double getValueAsDouble(ConstantFP *Op) {
   APF.convert(APFloat::IEEEdouble, APFloat::rmNearestTiesToEven, &unused);
   return APF.convertToDouble();
 }
 
 Constant *ConstantFoldScalarCall(StringRef Name, unsigned IntrinsicID, Type *Ty,
                                  ArrayRef<Constant *> Operands,
                                  const TargetLibraryInfo *TLI) {
   if (Operands.size() == 1) {
+    if (UndefValue *Op = dyn_cast<UndefValue>(Operands[0])) {
+      // cosine(arg) is between -1 and 1. cosine(invalid arg) is NaN
+      if (IntrinsicID == Intrinsic::cos)
+        return Constant::getNullValue(Ty);
+    }
     if (ConstantFP *Op = dyn_cast<ConstantFP>(Operands[0])) {
       if (IntrinsicID == Intrinsic::convert_to_fp16) {
         APFloat Val(Op->getValueAPF());
 
         bool lost = false;
         Val.convert(APFloat::IEEEhalf, APFloat::rmNearestTiesToEven, &lost);
 
         return ConstantInt::get(Ty->getContext(), Val.bitcastToAPInt());

test/Transforms/InstCombine/cos_intrinsic.ll

+; RUN: opt < %s -instcombine -S | FileCheck %s
+; This test makes sure that the undef is propagated for the cos instrinsic
+
+declare double    @llvm.cos.f64(double %Val)
+declare float     @llvm.cos.f32(float %Val)
+
+; Function Attrs: nounwind readnone 
+define i32 @test1() {
+; CHECK-LABEL: define i32 @test1(
+; CHECK-NEXT: ret i32 0 
+  %1 = call double @llvm.cos.f64(double undef) 
+  %2 = fcmp ogt double %1, 0.000000e+00
+  %3 = zext i1 %2 to i32
+  ret i32 %3
+}
+
+
+; Function Attrs: nounwind readnone 
+define float @test2(float %d) {
+; CHECK-LABEL: define float @test2(
+; CHECK-NEXT: %cosval = call float @llvm.cos.f32(float %d) 
+   %cosval   = call float @llvm.cos.f32(float %d)
+   %cosval2  = call float @llvm.cos.f32(float undef)
+   %fsum   = fadd float %cosval2, %cosval
+   ret float %fsum
+; CHECK-NEXT: %fsum
+; CHECK: ret float %fsum 
+}