Improve reduction intrinsics by overloading result value.

This patch uses the mechanism from D62995 to strengthen the
definitions of the reduction intrinsics by letting the scalar
result/accumulator type be overloaded from the vector element type.

For example:

  ; The LLVM LangRef specifies that the scalar result must equal the
  ; vector element type, but this is not checked/enforced by LLVM.
  declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> %a)

This patch changes that into:

  declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> %a)

Which has the type-constraint more explicit and causes LLVM to check
the result type with the vector element type.

Reviewers: RKSimon, arsenm, rnk, greened, aemerson

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D62996


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@363240 91177308-0d34-0410-b5e6-96231b3b80d8

docs/LangRef.rst

@@ -13719,8 +13719,8 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %a)
-      declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> %a)
+      declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %a)
+      declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> %a)
 
 Overview:
 """""""""
@@ -13780,8 +13780,8 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> %a)
-      declare i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> %a)
+      declare i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> %a)
+      declare i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> %a)
 
 Overview:
 """""""""
@@ -13840,7 +13840,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13861,7 +13861,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13882,7 +13882,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13903,7 +13903,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13924,7 +13924,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13945,7 +13945,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13966,7 +13966,7 @@ Syntax:
 
 ::
 
-      declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %a)
+      declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> %a)
 
 Overview:
 """""""""
@@ -13987,8 +13987,8 @@ Syntax:
 
 ::
 
-      declare float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %a)
-      declare double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double> %a)
+      declare float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float> %a)
+      declare double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double> %a)
 
 Overview:
 """""""""
@@ -14012,8 +14012,8 @@ Syntax:
 
 ::
 
-      declare float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %a)
-      declare double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double> %a)
+      declare float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float> %a)
+      declare double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double> %a)
 
 Overview:
 """""""""

include/llvm/IR/Intrinsics.h

@@ -99,7 +99,8 @@ namespace Intrinsic {
       Void, VarArg, MMX, Token, Metadata, Half, Float, Double, Quad,
       Integer, Vector, Pointer, Struct,
       Argument, ExtendArgument, TruncArgument, HalfVecArgument,
-      SameVecWidthArgument, PtrToArgument, PtrToElt, VecOfAnyPtrsToElt
+      SameVecWidthArgument, PtrToArgument, PtrToElt, VecOfAnyPtrsToElt,
+      VecElementArgument
     } Kind;
 
     union {
@@ -124,13 +125,14 @@ namespace Intrinsic {
       assert(Kind == Argument || Kind == ExtendArgument ||
              Kind == TruncArgument || Kind == HalfVecArgument ||
              Kind == SameVecWidthArgument || Kind == PtrToArgument ||
-             Kind == PtrToElt);
+             Kind == PtrToElt || Kind == VecElementArgument);
       return Argument_Info >> 3;
     }
     ArgKind getArgumentKind() const {
       assert(Kind == Argument || Kind == ExtendArgument ||
              Kind == TruncArgument || Kind == HalfVecArgument ||
-             Kind == SameVecWidthArgument || Kind == PtrToArgument);
+             Kind == SameVecWidthArgument || Kind == PtrToArgument ||
+             Kind == VecElementArgument);
       return (ArgKind)(Argument_Info & 7);
     }
 

include/llvm/IR/Intrinsics.td

@@ -173,6 +173,7 @@ class LLVMScalarOrSameVectorWidth<int idx, LLVMType elty>
 class LLVMPointerTo<int num> : LLVMMatchType<num>;
 class LLVMPointerToElt<int num> : LLVMMatchType<num>;
 class LLVMVectorOfAnyPointersToElt<int num> : LLVMMatchType<num>;
+class LLVMVectorElementType<int num> : LLVMMatchType<num>;
 
 // Match the type of another intrinsic parameter that is expected to be a
 // vector type, but change the element count to be half as many
@@ -1148,37 +1149,37 @@ def int_experimental_vector_reduce_v2_fmul : Intrinsic<[llvm_anyfloat_ty],
                                                        [LLVMMatchType<0>,
                                                         llvm_anyvector_ty],
                                                        [IntrNoMem]>;
-def int_experimental_vector_reduce_add : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_add : Intrinsic<[LLVMVectorElementType<0>],
                                                    [llvm_anyvector_ty],
                                                    [IntrNoMem]>;
-def int_experimental_vector_reduce_mul : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_mul : Intrinsic<[LLVMVectorElementType<0>],
                                                    [llvm_anyvector_ty],
                                                    [IntrNoMem]>;
-def int_experimental_vector_reduce_and : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_and : Intrinsic<[LLVMVectorElementType<0>],
                                                    [llvm_anyvector_ty],
                                                    [IntrNoMem]>;
-def int_experimental_vector_reduce_or : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_or : Intrinsic<[LLVMVectorElementType<0>],
                                                   [llvm_anyvector_ty],
                                                   [IntrNoMem]>;
-def int_experimental_vector_reduce_xor : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_xor : Intrinsic<[LLVMVectorElementType<0>],
                                                    [llvm_anyvector_ty],
                                                    [IntrNoMem]>;
-def int_experimental_vector_reduce_smax : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_smax : Intrinsic<[LLVMVectorElementType<0>],
                                                     [llvm_anyvector_ty],
                                                     [IntrNoMem]>;
-def int_experimental_vector_reduce_smin : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_smin : Intrinsic<[LLVMVectorElementType<0>],
                                                     [llvm_anyvector_ty],
                                                     [IntrNoMem]>;
-def int_experimental_vector_reduce_umax : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_umax : Intrinsic<[LLVMVectorElementType<0>],
                                                     [llvm_anyvector_ty],
                                                     [IntrNoMem]>;
-def int_experimental_vector_reduce_umin : Intrinsic<[llvm_anyint_ty],
+def int_experimental_vector_reduce_umin : Intrinsic<[LLVMVectorElementType<0>],
                                                     [llvm_anyvector_ty],
                                                     [IntrNoMem]>;
-def int_experimental_vector_reduce_fmax : Intrinsic<[llvm_anyfloat_ty],
+def int_experimental_vector_reduce_fmax : Intrinsic<[LLVMVectorElementType<0>],
                                                     [llvm_anyvector_ty],
                                                     [IntrNoMem]>;
-def int_experimental_vector_reduce_fmin : Intrinsic<[llvm_anyfloat_ty],
+def int_experimental_vector_reduce_fmin : Intrinsic<[LLVMVectorElementType<0>],
                                                     [llvm_anyvector_ty],
                                                     [IntrNoMem]>;
 

lib/IR/Function.cpp

@@ -700,7 +700,8 @@ enum IIT_Info {
   IIT_STRUCT6 = 38,
   IIT_STRUCT7 = 39,
   IIT_STRUCT8 = 40,
-  IIT_F128 = 41
+  IIT_F128 = 41,
+  IIT_VEC_ELEMENT = 42
 };
 
 static void DecodeIITType(unsigned &NextElt, ArrayRef<unsigned char> Infos,
@@ -865,6 +866,12 @@ static void DecodeIITType(unsigned &NextElt, ArrayRef<unsigned char> Infos,
       DecodeIITType(NextElt, Infos, OutputTable);
     return;
   }
+  case IIT_VEC_ELEMENT: {
+    unsigned ArgInfo = (NextElt == Infos.size() ? 0 : Infos[NextElt++]);
+    OutputTable.push_back(IITDescriptor::get(IITDescriptor::VecElementArgument,
+                                             ArgInfo));
+    return;
+  }
   }
   llvm_unreachable("unhandled");
 }
@@ -977,6 +984,12 @@ static Type *DecodeFixedType(ArrayRef<Intrinsic::IITDescriptor> &Infos,
     Type *EltTy = VTy->getVectorElementType();
     return PointerType::getUnqual(EltTy);
   }
+  case IITDescriptor::VecElementArgument: {
+    Type *Ty = Tys[D.getArgumentNumber()];
+    if (VectorType *VTy = dyn_cast<VectorType>(Ty))
+      return VTy->getElementType();
+    llvm_unreachable("Expected an argument of Vector Type");
+  }
   case IITDescriptor::VecOfAnyPtrsToElt:
     // Return the overloaded type (which determines the pointers address space)
     return Tys[D.getOverloadArgNumber()];
@@ -1237,6 +1250,12 @@ static bool matchIntrinsicType(
       return ThisArgEltTy->getElementType() !=
              ReferenceType->getVectorElementType();
     }
+    case IITDescriptor::VecElementArgument: {
+      if (D.getArgumentNumber() >= ArgTys.size())
+        return IsDeferredCheck ? true : DeferCheck(Ty);
+      auto *ReferenceType = dyn_cast<VectorType>(ArgTys[D.getArgumentNumber()]);
+      return !ReferenceType || Ty != ReferenceType->getElementType();
+    }
   }
   llvm_unreachable("unhandled");
 }

lib/IR/IRBuilder.cpp

@@ -313,7 +313,7 @@ static CallInst *getReductionIntrinsic(IRBuilderBase *Builder, Intrinsic::ID ID,
                                     Value *Src) {
   Module *M = Builder->GetInsertBlock()->getParent()->getParent();
   Value *Ops[] = {Src};
-  Type *Tys[] = { Src->getType()->getVectorElementType(), Src->getType() };
+  Type *Tys[] = { Src->getType() };
   auto Decl = Intrinsic::getDeclaration(M, ID, Tys);
   return createCallHelper(Decl, Ops, Builder);
 }

test/Analysis/CostModel/AArch64/vector-reduce.ll

@@ -2,278 +2,278 @@
 ; RUN: llc < %s -mtriple=aarch64--linux-gnu | FileCheck %s --check-prefix=CODE
 
 ; COST-LABEL: add.i8.v8i8
-; COST:       Found an estimated cost of 1 for instruction: %r = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> %v)
+; COST:       Found an estimated cost of 1 for instruction: %r = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> %v)
 ; CODE-LABEL: add.i8.v8i8
 ; CODE:       addv b0, v0.8b
 define i8 @add.i8.v8i8(<8 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: add.i8.v16i8
-; COST:       Found an estimated cost of 1 for instruction: %r = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> %v)
+; COST:       Found an estimated cost of 1 for instruction: %r = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> %v)
 ; CODE-LABEL: add.i8.v16i8
 ; CODE:       addv b0, v0.16b
 define i8 @add.i8.v16i8(<16 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: add.i16.v4i16
-; COST:       Found an estimated cost of 1 for instruction: %r = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> %v)
+; COST:       Found an estimated cost of 1 for instruction: %r = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> %v)
 ; CODE-LABEL: add.i16.v4i16
 ; CODE:       addv h0, v0.4h
 define i16 @add.i16.v4i16(<4 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: add.i16.v8i16
-; COST:       Found an estimated cost of 1 for instruction: %r = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> %v)
+; COST:       Found an estimated cost of 1 for instruction: %r = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> %v)
 ; CODE-LABEL: add.i16.v8i16
 ; CODE:       addv h0, v0.8h
 define i16 @add.i16.v8i16(<8 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: add.i32.v4i32
-; COST:       Found an estimated cost of 1 for instruction: %r = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %v)
+; COST:       Found an estimated cost of 1 for instruction: %r = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v)
 ; CODE-LABEL: add.i32.v4i32
 ; CODE:       addv s0, v0.4s
 define i32 @add.i32.v4i32(<4 x i32> %v) {
-  %r = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %v)
+  %r = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %v)
   ret i32 %r
 }
 
 ; COST-LABEL: umin.i8.v8i8
-; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> %v)
 ; CODE-LABEL: umin.i8.v8i8
 ; CODE:       uminv b0, v0.8b
 define i8 @umin.i8.v8i8(<8 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: umin.i8.v16i8
-; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> %v)
+; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> %v)
 ; CODE-LABEL: umin.i8.v16i8
 ; CODE:       uminv b0, v0.16b
 define i8 @umin.i8.v16i8(<16 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: umin.i16.v4i16
-; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> %v)
+; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> %v)
 ; CODE-LABEL: umin.i16.v4i16
 ; CODE:       uminv h0, v0.4h
 define i16 @umin.i16.v4i16(<4 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: umin.i16.v8i16
-; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> %v)
 ; CODE-LABEL: umin.i16.v8i16
 ; CODE:       uminv h0, v0.8h
 define i16 @umin.i16.v8i16(<8 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: umin.i32.v4i32
-; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %v)
+; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> %v)
 ; CODE-LABEL: umin.i32.v4i32
 ; CODE:       uminv s0, v0.4s
 define i32 @umin.i32.v4i32(<4 x i32> %v) {
-  %r = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %v)
+  %r = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> %v)
   ret i32 %r
 }
 
 ; COST-LABEL: umax.i8.v8i8
-; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> %v)
 ; CODE-LABEL: umax.i8.v8i8
 ; CODE:       umaxv b0, v0.8b
 define i8 @umax.i8.v8i8(<8 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: umax.i8.v16i8
-; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> %v)
+; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> %v)
 ; CODE-LABEL: umax.i8.v16i8
 ; CODE:       umaxv b0, v0.16b
 define i8 @umax.i8.v16i8(<16 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: umax.i16.v4i16
-; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> %v)
+; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> %v)
 ; CODE-LABEL: umax.i16.v4i16
 ; CODE:       umaxv h0, v0.4h
 define i16 @umax.i16.v4i16(<4 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: umax.i16.v8i16
-; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> %v)
 ; CODE-LABEL: umax.i16.v8i16
 ; CODE:       umaxv h0, v0.8h
 define i16 @umax.i16.v8i16(<8 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: umax.i32.v4i32
-; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %v)
+; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> %v)
 ; CODE-LABEL: umax.i32.v4i32
 ; CODE:       umaxv s0, v0.4s
 define i32 @umax.i32.v4i32(<4 x i32> %v) {
-  %r = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %v)
+  %r = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> %v)
   ret i32 %r
 }
 
 ; COST-LABEL: smin.i8.v8i8
-; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> %v)
 ; CODE-LABEL: smin.i8.v8i8
 ; CODE:       sminv b0, v0.8b
 define i8 @smin.i8.v8i8(<8 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: smin.i8.v16i8
-; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> %v)
+; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> %v)
 ; CODE-LABEL: smin.i8.v16i8
 ; CODE:       sminv b0, v0.16b
 define i8 @smin.i8.v16i8(<16 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: smin.i16.v4i16
-; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> %v)
+; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> %v)
 ; CODE-LABEL: smin.i16.v4i16
 ; CODE:       sminv h0, v0.4h
 define i16 @smin.i16.v4i16(<4 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: smin.i16.v8i16
-; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> %v)
 ; CODE-LABEL: smin.i16.v8i16
 ; CODE:       sminv h0, v0.8h
 define i16 @smin.i16.v8i16(<8 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: smin.i32.v4i32
-; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %v)
+; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> %v)
 ; CODE-LABEL: smin.i32.v4i32
 ; CODE:       sminv s0, v0.4s
 define i32 @smin.i32.v4i32(<4 x i32> %v) {
-  %r = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %v)
+  %r = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> %v)
   ret i32 %r
 }
 
 ; COST-LABEL: smax.i8.v8i8
-; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> %v)
 ; CODE-LABEL: smax.i8.v8i8
 ; CODE:       smaxv b0, v0.8b
 define i8 @smax.i8.v8i8(<8 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: smax.i8.v16i8
-; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> %v)
+; COST:       Found an estimated cost of 608 for instruction: %r = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> %v)
 ; CODE-LABEL: smax.i8.v16i8
 ; CODE:       smaxv b0, v0.16b
 define i8 @smax.i8.v16i8(<16 x i8> %v) {
-  %r = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> %v)
+  %r = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> %v)
   ret i8 %r
 }
 
 ; COST-LABEL: smax.i16.v4i16
-; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> %v)
+; COST:       Found an estimated cost of 64 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> %v)
 ; CODE-LABEL: smax.i16.v4i16
 ; CODE:       smaxv h0, v0.4h
 define i16 @smax.i16.v4i16(<4 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: smax.i16.v8i16
-; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> %v)
+; COST:       Found an estimated cost of 216 for instruction: %r = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> %v)
 ; CODE-LABEL: smax.i16.v8i16
 ; CODE:       smaxv h0, v0.8h
 define i16 @smax.i16.v8i16(<8 x i16> %v) {
-  %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> %v)
+  %r = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> %v)
   ret i16 %r
 }
 
 ; COST-LABEL: smax.i32.v4i32
-; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %v)
+; COST:       Found an estimated cost of 34 for instruction: %r = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> %v)
 ; CODE-LABEL: smax.i32.v4i32
 ; CODE:       smaxv s0, v0.4s
 define i32 @smax.i32.v4i32(<4 x i32> %v) {
-  %r = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %v)
+  %r = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> %v)
   ret i32 %r
 }
 
 ; COST-LABEL: fmin.f32.v4f32
-; COST:       Found an estimated cost of 34 for instruction: %r = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %v)
+; COST:       Found an estimated cost of 34 for instruction: %r = call nnan float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float> %v)
 ; CODE-LABEL: fmin.f32.v4f32
 ; CODE:       fminnmv s0, v0.4s
 define float @fmin.f32.v4f32(<4 x float> %v) {
-  %r = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %v)
+  %r = call nnan float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float> %v)
   ret float %r
 }
 
 ; COST-LABEL: fmax.f32.v4f32
-; COST:       Found an estimated cost of 34 for instruction: %r = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %v)
+; COST:       Found an estimated cost of 34 for instruction: %r = call nnan float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float> %v)
 ; CODE-LABEL: fmax.f32.v4f32
 ; CODE:       fmaxnmv s0, v0.4s
 define float @fmax.f32.v4f32(<4 x float> %v) {
-  %r = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %v)
+  %r = call nnan float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float> %v)
   ret float %r
 }
 
-declare i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>)
-
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>)
-
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>)
-
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>)
-
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>)
-
-declare float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float>)
-
-declare float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float>)
+declare i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+
+declare i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32>)
+
+declare i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32>)
+
+declare i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32>)
+
+declare i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32>)
+
+declare float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float>)
+
+declare float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float>)

test/Analysis/CostModel/X86/reduce-add-widen.ll

@@ -10,298 +10,298 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX-LABEL: 'reduce_i64'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX-LABEL: 'reduce_i32'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 73 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 73 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-add.ll

@@ -10,298 +10,298 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX-LABEL: 'reduce_i64'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX-LABEL: 'reduce_i32'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; AVX-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; AVX-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 73 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 73 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-and-widen.ll

@@ -10,369 +10,369 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE-LABEL: 'reduce_i32'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i1(i32 %arg) {
 ; SSE-LABEL: 'reduce_i1'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i1'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i1'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i1'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i1'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i1'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1   = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-  %V2   = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-  %V4   = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-  %V8   = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-  %V16  = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-  %V32  = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-  %V64  = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-  %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+  %V1   = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+  %V2   = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+  %V4   = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+  %V8   = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+  %V16  = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+  %V32  = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+  %V64  = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+  %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8>)
 
-declare i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1>)

test/Analysis/CostModel/X86/reduce-and.ll

@@ -10,369 +10,369 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE-LABEL: 'reduce_i32'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i1(i32 %arg) {
 ; SSE-LABEL: 'reduce_i1'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i1'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i1'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i1'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i1'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i1'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1   = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> undef)
-  %V2   = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> undef)
-  %V4   = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> undef)
-  %V8   = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> undef)
-  %V16  = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> undef)
-  %V32  = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> undef)
-  %V64  = call i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1> undef)
-  %V128 = call i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1> undef)
+  %V1   = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> undef)
+  %V2   = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> undef)
+  %V4   = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> undef)
+  %V8   = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> undef)
+  %V16  = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> undef)
+  %V32  = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> undef)
+  %V64  = call i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1> undef)
+  %V128 = call i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8>)
 
-declare i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v64i1(<64 x i1>)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v128i1(<128 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v64i1(<64 x i1>)
+declare i1 @llvm.experimental.vector.reduce.and.v128i1(<128 x i1>)

test/Analysis/CostModel/X86/reduce-mul-widen.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 66 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 66 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 41 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 59 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 95 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 41 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 59 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 95 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i64'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i64'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i64'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 101 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 173 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 101 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 173 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 171 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 197 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 249 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 171 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 197 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 249 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 106 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 123 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 157 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 106 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 123 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 157 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 115 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 115 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-mul.ll

@@ -10,330 +10,330 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 66 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 66 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 41 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 59 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 95 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 41 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 59 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 95 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i64'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i64'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i64'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 57 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i32'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i32'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i32'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 101 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 173 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 101 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 173 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 89 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 137 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 171 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 197 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 249 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 171 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 197 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 249 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 106 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 123 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 157 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 106 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 123 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 157 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 115 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 115 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 125 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-or-widen.ll

@@ -10,369 +10,369 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE-LABEL: 'reduce_i32'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i1(i32 %arg) {
 ; SSE-LABEL: 'reduce_i1'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i1'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i1'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i1'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i1'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i1'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1   = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-  %V2   = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-  %V4   = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-  %V8   = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-  %V16  = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-  %V32  = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-  %V64  = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-  %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+  %V1   = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+  %V2   = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+  %V4   = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+  %V8   = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+  %V16  = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+  %V32  = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+  %V64  = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+  %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8>)
 
-declare i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1>)

test/Analysis/CostModel/X86/reduce-or.ll

@@ -10,369 +10,369 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE-LABEL: 'reduce_i32'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i1(i32 %arg) {
 ; SSE-LABEL: 'reduce_i1'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i1'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i1'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i1'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i1'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i1'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1   = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> undef)
-  %V2   = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> undef)
-  %V4   = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> undef)
-  %V8   = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> undef)
-  %V16  = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> undef)
-  %V32  = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> undef)
-  %V64  = call i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1> undef)
-  %V128 = call i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1> undef)
+  %V1   = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> undef)
+  %V2   = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> undef)
+  %V4   = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> undef)
+  %V8   = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> undef)
+  %V16  = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> undef)
+  %V32  = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> undef)
+  %V64  = call i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1> undef)
+  %V128 = call i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.or.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8>)
 
-declare i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v64i1(<64 x i1>)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v128i1(<128 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v64i1(<64 x i1>)
+declare i1 @llvm.experimental.vector.reduce.or.v128i1(<128 x i1>)

test/Analysis/CostModel/X86/reduce-smax-widen.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-smax.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-smin-widen.ll

@@ -10,305 +10,305 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V4  = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-smin.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-umax-widen.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-umax.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-umin-widen.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-umin.ll

@@ -10,314 +10,314 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i64'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i64'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i64'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i32'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i32'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i32'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8>)

test/Analysis/CostModel/X86/reduce-xor-widen.ll

@@ -10,391 +10,391 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE-LABEL: 'reduce_i32'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i1(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i1'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i1'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i1'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i1'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i1'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i1'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i1'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i1'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1   = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-  %V2   = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-  %V4   = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-  %V8   = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-  %V16  = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-  %V32  = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-  %V64  = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-  %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+  %V1   = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+  %V2   = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+  %V4   = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+  %V8   = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+  %V16  = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+  %V32  = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+  %V64  = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+  %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8>)
 
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1>)

test/Analysis/CostModel/X86/reduce-xor.ll

@@ -10,391 +10,391 @@
 
 define i32 @reduce_i64(i32 %arg) {
 ; SSE-LABEL: 'reduce_i64'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i64'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i64'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i64'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1  = call i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64> undef)
-  %V2  = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> undef)
-  %V4  = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> undef)
-  %V8  = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> undef)
-  %V16 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> undef)
+  %V1  = call i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64> undef)
+  %V2  = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> undef)
+  %V4  = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> undef)
+  %V8  = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> undef)
+  %V16 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i32(i32 %arg) {
 ; SSE-LABEL: 'reduce_i32'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i32'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i32'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512-LABEL: 'reduce_i32'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
 ; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> undef)
-  %V4  = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> undef)
-  %V8  = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> undef)
-  %V16 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> undef)
-  %V32 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> undef)
+  %V2  = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> undef)
+  %V4  = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> undef)
+  %V8  = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> undef)
+  %V16 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> undef)
+  %V32 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i16(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i16'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i16'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i16'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i16'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i16'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i16'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i16'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i16'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2  = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> undef)
-  %V4  = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> undef)
-  %V8  = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> undef)
-  %V16 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> undef)
-  %V32 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> undef)
-  %V64 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> undef)
+  %V2  = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> undef)
+  %V4  = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> undef)
+  %V8  = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> undef)
+  %V16 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> undef)
+  %V32 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> undef)
+  %V64 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i8(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i8'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i8'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i8'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i8'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i8'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i8'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i8'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 55 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i8'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V2   = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> undef)
-  %V4   = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> undef)
-  %V8   = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> undef)
-  %V16  = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> undef)
-  %V32  = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> undef)
-  %V64  = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> undef)
-  %V128 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> undef)
+  %V2   = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> undef)
+  %V4   = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> undef)
+  %V8   = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> undef)
+  %V16  = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> undef)
+  %V32  = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> undef)
+  %V64  = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> undef)
+  %V128 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> undef)
   ret i32 undef
 }
 
 define i32 @reduce_i1(i32 %arg) {
 ; SSE2-LABEL: 'reduce_i1'
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSSE3-LABEL: 'reduce_i1'
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SSE42-LABEL: 'reduce_i1'
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX1-LABEL: 'reduce_i1'
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 49 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX2-LABEL: 'reduce_i1'
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512F-LABEL: 'reduce_i1'
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512BW-LABEL: 'reduce_i1'
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 326 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 775 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 776 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; AVX512DQ-LABEL: 'reduce_i1'
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V4 = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V8 = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V16 = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 134 for instruction: %V32 = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %V64 = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
 ; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
-  %V1   = call i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1> undef)
-  %V2   = call i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1> undef)
-  %V4   = call i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1> undef)
-  %V8   = call i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1> undef)
-  %V16  = call i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1> undef)
-  %V32  = call i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1> undef)
-  %V64  = call i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1> undef)
-  %V128 = call i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1> undef)
+  %V1   = call i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1> undef)
+  %V2   = call i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1> undef)
+  %V4   = call i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1> undef)
+  %V8   = call i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1> undef)
+  %V16  = call i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1> undef)
+  %V32  = call i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1> undef)
+  %V64  = call i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1> undef)
+  %V128 = call i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1> undef)
   ret i32 undef
 }
 
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v1i64(<1 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v1i64(<1 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8>)
 
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v1i1(<1 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v2i1(<2 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v4i1(<4 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v8i1(<8 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v16i1(<16 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v32i1(<32 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v64i1(<64 x i1>)
-declare i1 @llvm.experimental.vector.reduce.xor.i1.v128i1(<128 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v1i1(<1 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v2i1(<2 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v4i1(<4 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v8i1(<8 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v16i1(<16 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v32i1(<32 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v64i1(<64 x i1>)
+declare i1 @llvm.experimental.vector.reduce.xor.v128i1(<128 x i1>)

test/CodeGen/AArch64/aarch64-addv.ll

@@ -1,16 +1,16 @@
 ; RUN: llc < %s -mtriple=aarch64-eabi -aarch64-neon-syntax=generic | FileCheck %s
 
 ; Function Attrs: nounwind readnone
-declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>)
+declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64>)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+declare i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>)
+declare i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>)
 
 define i8 @add_B(<16 x i8>* %arr)  {
 ; CHECK-LABEL: add_B
 ; CHECK: addv {{b[0-9]+}}, {{v[0-9]+}}.16b
   %bin.rdx = load <16 x i8>, <16 x i8>* %arr
-  %r = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> %bin.rdx)
+  %r = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> %bin.rdx)
   ret i8 %r
 }
 
@@ -18,7 +18,7 @@ define i16 @add_H(<8 x i16>* %arr)  {
 ; CHECK-LABEL: add_H
 ; CHECK: addv {{h[0-9]+}}, {{v[0-9]+}}.8h
   %bin.rdx = load <8 x i16>, <8 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> %bin.rdx)
+  %r = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> %bin.rdx)
   ret i16 %r
 }
 
@@ -26,7 +26,7 @@ define i32 @add_S( <4 x i32>* %arr)  {
 ; CHECK-LABEL: add_S
 ; CHECK: addv {{s[0-9]+}}, {{v[0-9]+}}.4s
   %bin.rdx = load <4 x i32>, <4 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %bin.rdx)
+  %r = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %bin.rdx)
   ret i32 %r
 }
 
@@ -34,11 +34,11 @@ define i64 @add_D(<2 x i64>* %arr)  {
 ; CHECK-LABEL: add_D
 ; CHECK-NOT: addv
   %bin.rdx = load <2 x i64>, <2 x i64>* %arr
-  %r = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> %bin.rdx)
+  %r = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> %bin.rdx)
   ret i64 %r
 }
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
 
 define i32 @oversized_ADDV_256(i8* noalias nocapture readonly %arg1, i8* noalias nocapture readonly %arg2) {
 ; CHECK-LABEL: oversized_ADDV_256
@@ -54,16 +54,16 @@ entry:
   %7 = icmp slt <8 x i32> %6, zeroinitializer
   %8 = sub nsw <8 x i32> zeroinitializer, %6
   %9 = select <8 x i1> %7, <8 x i32> %8, <8 x i32> %6
-  %r = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> %9)
+  %r = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> %9)
   ret i32 %r
 }
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32>)
 
 define i32 @oversized_ADDV_512(<16 x i32>* %arr)  {
 ; CHECK-LABEL: oversized_ADDV_512
 ; CHECK: addv {{s[0-9]+}}, {{v[0-9]+}}.4s
   %bin.rdx = load <16 x i32>, <16 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> %bin.rdx)
+  %r = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> %bin.rdx)
   ret i32 %r
 }

test/CodeGen/AArch64/aarch64-minmaxv.ll

@@ -2,28 +2,28 @@
 
 target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
 
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>)
-
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>)
-
-declare float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float>)
+declare i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32>)
+declare i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32>)
+
+declare i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32>)
+declare i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8>)
+declare i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16>)
+declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32>)
+
+declare float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float>)
 
 ; CHECK-LABEL: smax_B
 ; CHECK: smaxv {{b[0-9]+}}, {{v[0-9]+}}.16b
 define i8 @smax_B(<16 x i8>* nocapture readonly %arr)  {
   %arr.load = load <16 x i8>, <16 x i8>* %arr
-  %r = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> %arr.load)
+  %r = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> %arr.load)
   ret i8 %r
 }
 
@@ -31,7 +31,7 @@ define i8 @smax_B(<16 x i8>* nocapture readonly %arr)  {
 ; CHECK: smaxv {{h[0-9]+}}, {{v[0-9]+}}.8h
 define i16 @smax_H(<8 x i16>* nocapture readonly %arr) {
   %arr.load = load <8 x i16>, <8 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> %arr.load)
   ret i16 %r
 }
 
@@ -39,7 +39,7 @@ define i16 @smax_H(<8 x i16>* nocapture readonly %arr) {
 ; CHECK: smaxv {{s[0-9]+}}, {{v[0-9]+}}.4s
 define i32 @smax_S(<4 x i32> * nocapture readonly %arr)  {
   %arr.load = load <4 x i32>, <4 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> %arr.load)
   ret i32 %r
 }
 
@@ -47,7 +47,7 @@ define i32 @smax_S(<4 x i32> * nocapture readonly %arr)  {
 ; CHECK: umaxv {{b[0-9]+}}, {{v[0-9]+}}.16b
 define i8 @umax_B(<16 x i8>* nocapture readonly %arr)  {
   %arr.load = load <16 x i8>, <16 x i8>* %arr
-  %r = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> %arr.load)
+  %r = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> %arr.load)
   ret i8 %r
 }
 
@@ -55,7 +55,7 @@ define i8 @umax_B(<16 x i8>* nocapture readonly %arr)  {
 ; CHECK: umaxv {{h[0-9]+}}, {{v[0-9]+}}.8h
 define i16 @umax_H(<8 x i16>* nocapture readonly %arr)  {
   %arr.load = load <8 x i16>, <8 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> %arr.load)
   ret i16 %r
 }
 
@@ -63,7 +63,7 @@ define i16 @umax_H(<8 x i16>* nocapture readonly %arr)  {
 ; CHECK: umaxv {{s[0-9]+}}, {{v[0-9]+}}.4s
 define i32 @umax_S(<4 x i32>* nocapture readonly %arr) {
   %arr.load = load <4 x i32>, <4 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> %arr.load)
   ret i32 %r
 }
 
@@ -71,7 +71,7 @@ define i32 @umax_S(<4 x i32>* nocapture readonly %arr) {
 ; CHECK: sminv {{b[0-9]+}}, {{v[0-9]+}}.16b
 define i8 @smin_B(<16 x i8>* nocapture readonly %arr) {
   %arr.load = load <16 x i8>, <16 x i8>* %arr
-  %r = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> %arr.load)
+  %r = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> %arr.load)
   ret i8 %r
 }
 
@@ -79,7 +79,7 @@ define i8 @smin_B(<16 x i8>* nocapture readonly %arr) {
 ; CHECK: sminv {{h[0-9]+}}, {{v[0-9]+}}.8h
 define i16 @smin_H(<8 x i16>* nocapture readonly %arr) {
   %arr.load = load <8 x i16>, <8 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> %arr.load)
   ret i16 %r
 }
 
@@ -87,7 +87,7 @@ define i16 @smin_H(<8 x i16>* nocapture readonly %arr) {
 ; CHECK: sminv {{s[0-9]+}}, {{v[0-9]+}}.4s
 define i32 @smin_S(<4 x i32>* nocapture readonly %arr) {
   %arr.load = load <4 x i32>, <4 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> %arr.load)
   ret i32 %r
 }
 
@@ -95,7 +95,7 @@ define i32 @smin_S(<4 x i32>* nocapture readonly %arr) {
 ; CHECK: uminv {{b[0-9]+}}, {{v[0-9]+}}.16b
 define i8 @umin_B(<16 x i8>* nocapture readonly %arr)  {
   %arr.load = load <16 x i8>, <16 x i8>* %arr
-  %r = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> %arr.load)
+  %r = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> %arr.load)
   ret i8 %r
 }
 
@@ -103,7 +103,7 @@ define i8 @umin_B(<16 x i8>* nocapture readonly %arr)  {
 ; CHECK: uminv {{h[0-9]+}}, {{v[0-9]+}}.8h
 define i16 @umin_H(<8 x i16>* nocapture readonly %arr)  {
   %arr.load = load <8 x i16>, <8 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> %arr.load)
   ret i16 %r
 }
 
@@ -111,7 +111,7 @@ define i16 @umin_H(<8 x i16>* nocapture readonly %arr)  {
 ; CHECK: uminv {{s[0-9]+}}, {{v[0-9]+}}.4s
 define i32 @umin_S(<4 x i32>* nocapture readonly %arr) {
   %arr.load = load <4 x i32>, <4 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> %arr.load)
   ret i32 %r
 }
 
@@ -119,7 +119,7 @@ define i32 @umin_S(<4 x i32>* nocapture readonly %arr) {
 ; CHECK: fmaxnmv
 define float @fmaxnm_S(<4 x float>* nocapture readonly %arr) {
   %arr.load  = load <4 x float>, <4 x float>* %arr
-  %r = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %arr.load)
+  %r = call nnan float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float> %arr.load)
   ret float %r
 }
 
@@ -127,22 +127,22 @@ define float @fmaxnm_S(<4 x float>* nocapture readonly %arr) {
 ; CHECK: fminnmv
 define float @fminnm_S(<4 x float>* nocapture readonly %arr) {
   %arr.load  = load <4 x float>, <4 x float>* %arr
-  %r = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %arr.load)
+  %r = call nnan float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float> %arr.load)
   ret float %r
 }
 
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16>)
 
 define i16 @oversized_umax_256(<16 x i16>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_umax_256
 ; CHECK: umax [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h
 ; CHECK: umaxv {{h[0-9]+}}, [[V0]]
   %arr.load = load <16 x i16>, <16 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> %arr.load)
   ret i16 %r
 }
 
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32>)
 
 define i32 @oversized_umax_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_umax_512
@@ -151,22 +151,22 @@ define i32 @oversized_umax_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-NEXT: umax [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
 ; CHECK-NEXT: umaxv {{s[0-9]+}}, [[V0]]
   %arr.load = load <16 x i32>, <16 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> %arr.load)
   ret i32 %r
 }
 
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16>)
 
 define i16 @oversized_umin_256(<16 x i16>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_umin_256
 ; CHECK: umin [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h
 ; CHECK: uminv {{h[0-9]+}}, [[V0]]
   %arr.load = load <16 x i16>, <16 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> %arr.load)
   ret i16 %r
 }
 
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32>)
 
 define i32 @oversized_umin_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_umin_512
@@ -175,22 +175,22 @@ define i32 @oversized_umin_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-NEXT: umin [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
 ; CHECK-NEXT: uminv {{s[0-9]+}}, [[V0]]
   %arr.load = load <16 x i32>, <16 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> %arr.load)
   ret i32 %r
 }
 
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16>)
 
 define i16 @oversized_smax_256(<16 x i16>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_smax_256
 ; CHECK: smax [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h
 ; CHECK: smaxv {{h[0-9]+}}, [[V0]]
   %arr.load = load <16 x i16>, <16 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> %arr.load)
   ret i16 %r
 }
 
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32>)
 
 define i32 @oversized_smax_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_smax_512
@@ -199,22 +199,22 @@ define i32 @oversized_smax_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-NEXT: smax [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
 ; CHECK-NEXT: smaxv {{s[0-9]+}}, [[V0]]
   %arr.load = load <16 x i32>, <16 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> %arr.load)
   ret i32 %r
 }
 
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16>)
 
 define i16 @oversized_smin_256(<16 x i16>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_smin_256
 ; CHECK: smin [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h
 ; CHECK: sminv {{h[0-9]+}}, [[V0]]
   %arr.load = load <16 x i16>, <16 x i16>* %arr
-  %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> %arr.load)
+  %r = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> %arr.load)
   ret i16 %r
 }
 
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32>)
 
 define i32 @oversized_smin_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-LABEL: oversized_smin_512
@@ -223,6 +223,6 @@ define i32 @oversized_smin_512(<16 x i32>* nocapture readonly %arr)  {
 ; CHECK-NEXT: smin [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
 ; CHECK-NEXT: sminv {{s[0-9]+}}, [[V0]]
   %arr.load = load <16 x i32>, <16 x i32>* %arr
-  %r = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> %arr.load)
+  %r = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> %arr.load)
   ret i32 %r
 }

test/CodeGen/AArch64/arm64-vabs.ll

@@ -141,7 +141,7 @@ define <2 x i64> @uabdl2_2d(<4 x i32>* %A, <4 x i32>* %B) nounwind {
   ret <2 x i64> %tmp4
 }
 
-declare i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16>)
 
 define i16 @uabdl8h_rdx(<16 x i8>* %a, <16 x i8>* %b) {
 ; CHECK-LABEL: uabdl8h_rdx
@@ -155,11 +155,11 @@ define i16 @uabdl8h_rdx(<16 x i8>* %a, <16 x i8>* %b) {
   %abcmp = icmp slt <16 x i16> %abdiff, zeroinitializer
   %ababs = sub nsw <16 x i16> zeroinitializer, %abdiff
   %absel = select <16 x i1> %abcmp, <16 x i16> %ababs, <16 x i16> %abdiff
-  %reduced_v = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> %absel)
+  %reduced_v = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> %absel)
   ret i16 %reduced_v
 }
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
 
 define i32 @uabdl4s_rdx(<8 x i16>* %a, <8 x i16>* %b) {
 ; CHECK-LABEL: uabdl4s_rdx
@@ -173,11 +173,11 @@ define i32 @uabdl4s_rdx(<8 x i16>* %a, <8 x i16>* %b) {
   %abcmp = icmp slt <8 x i32> %abdiff, zeroinitializer
   %ababs = sub nsw <8 x i32> zeroinitializer, %abdiff
   %absel = select <8 x i1> %abcmp, <8 x i32> %ababs, <8 x i32> %abdiff
-  %reduced_v = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> %absel)
+  %reduced_v = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> %absel)
   ret i32 %reduced_v
 }
 
-declare i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64>)
 
 define i64 @uabdl2d_rdx(<4 x i32>* %a, <4 x i32>* %b, i32 %h) {
 ; CHECK: uabdl2d_rdx
@@ -191,7 +191,7 @@ define i64 @uabdl2d_rdx(<4 x i32>* %a, <4 x i32>* %b, i32 %h) {
   %abcmp = icmp slt <4 x i64> %abdiff, zeroinitializer
   %ababs = sub nsw <4 x i64> zeroinitializer, %abdiff
   %absel = select <4 x i1> %abcmp, <4 x i64> %ababs, <4 x i64> %abdiff
-  %reduced_v = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> %absel)
+  %reduced_v = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> %absel)
   ret i64 %reduced_v
 }
 

test/CodeGen/AArch64/vecreduce-add-legalization.ll

@@ -1,28 +1,28 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s --check-prefix=CHECK
 
-declare i1 @llvm.experimental.vector.reduce.add.i1.v1i1(<1 x i1> %a)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v1i8(<1 x i8> %a)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v1i16(<1 x i16> %a)
-declare i24 @llvm.experimental.vector.reduce.add.i24.v1i24(<1 x i24> %a)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v1i32(<1 x i32> %a)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> %a)
-declare i128 @llvm.experimental.vector.reduce.add.i128.v1i128(<1 x i128> %a)
-
-declare i8 @llvm.experimental.vector.reduce.add.i8.v3i8(<3 x i8> %a)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v9i8(<9 x i8> %a)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v3i32(<3 x i32> %a)
-declare i1 @llvm.experimental.vector.reduce.add.i1.v4i1(<4 x i1> %a)
-declare i24 @llvm.experimental.vector.reduce.add.i24.v4i24(<4 x i24> %a)
-declare i128 @llvm.experimental.vector.reduce.add.i128.v2i128(<2 x i128> %a)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> %a)
+declare i1 @llvm.experimental.vector.reduce.add.v1i1(<1 x i1> %a)
+declare i8 @llvm.experimental.vector.reduce.add.v1i8(<1 x i8> %a)
+declare i16 @llvm.experimental.vector.reduce.add.v1i16(<1 x i16> %a)
+declare i24 @llvm.experimental.vector.reduce.add.v1i24(<1 x i24> %a)
+declare i32 @llvm.experimental.vector.reduce.add.v1i32(<1 x i32> %a)
+declare i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> %a)
+declare i128 @llvm.experimental.vector.reduce.add.v1i128(<1 x i128> %a)
+
+declare i8 @llvm.experimental.vector.reduce.add.v3i8(<3 x i8> %a)
+declare i8 @llvm.experimental.vector.reduce.add.v9i8(<9 x i8> %a)
+declare i32 @llvm.experimental.vector.reduce.add.v3i32(<3 x i32> %a)
+declare i1 @llvm.experimental.vector.reduce.add.v4i1(<4 x i1> %a)
+declare i24 @llvm.experimental.vector.reduce.add.v4i24(<4 x i24> %a)
+declare i128 @llvm.experimental.vector.reduce.add.v2i128(<2 x i128> %a)
+declare i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> %a)
 
 define i1 @test_v1i1(<1 x i1> %a) nounwind {
 ; CHECK-LABEL: test_v1i1:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    and w0, w0, #0x1
 ; CHECK-NEXT:    ret
-  %b = call i1 @llvm.experimental.vector.reduce.add.i1.v1i1(<1 x i1> %a)
+  %b = call i1 @llvm.experimental.vector.reduce.add.v1i1(<1 x i1> %a)
   ret i1 %b
 }
 
@@ -32,7 +32,7 @@ define i8 @test_v1i8(<1 x i8> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    umov w0, v0.b[0]
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.add.i8.v1i8(<1 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.add.v1i8(<1 x i8> %a)
   ret i8 %b
 }
 
@@ -42,7 +42,7 @@ define i16 @test_v1i16(<1 x i16> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    umov w0, v0.h[0]
 ; CHECK-NEXT:    ret
-  %b = call i16 @llvm.experimental.vector.reduce.add.i16.v1i16(<1 x i16> %a)
+  %b = call i16 @llvm.experimental.vector.reduce.add.v1i16(<1 x i16> %a)
   ret i16 %b
 }
 
@@ -50,7 +50,7 @@ define i24 @test_v1i24(<1 x i24> %a) nounwind {
 ; CHECK-LABEL: test_v1i24:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call i24 @llvm.experimental.vector.reduce.add.i24.v1i24(<1 x i24> %a)
+  %b = call i24 @llvm.experimental.vector.reduce.add.v1i24(<1 x i24> %a)
   ret i24 %b
 }
 
@@ -60,7 +60,7 @@ define i32 @test_v1i32(<1 x i32> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.add.i32.v1i32(<1 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.add.v1i32(<1 x i32> %a)
   ret i32 %b
 }
 
@@ -70,7 +70,7 @@ define i64 @test_v1i64(<1 x i64> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    fmov x0, d0
 ; CHECK-NEXT:    ret
-  %b = call i64 @llvm.experimental.vector.reduce.add.i64.v1i64(<1 x i64> %a)
+  %b = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> %a)
   ret i64 %b
 }
 
@@ -78,7 +78,7 @@ define i128 @test_v1i128(<1 x i128> %a) nounwind {
 ; CHECK-LABEL: test_v1i128:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call i128 @llvm.experimental.vector.reduce.add.i128.v1i128(<1 x i128> %a)
+  %b = call i128 @llvm.experimental.vector.reduce.add.v1i128(<1 x i128> %a)
   ret i128 %b
 }
 
@@ -92,7 +92,7 @@ define i8 @test_v3i8(<3 x i8> %a) nounwind {
 ; CHECK-NEXT:    addv h0, v0.4h
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.add.i8.v3i8(<3 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.add.v3i8(<3 x i8> %a)
   ret i8 %b
 }
 
@@ -109,7 +109,7 @@ define i8 @test_v9i8(<9 x i8> %a) nounwind {
 ; CHECK-NEXT:    addv b0, v0.16b
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.add.i8.v9i8(<9 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.add.v9i8(<9 x i8> %a)
   ret i8 %b
 }
 
@@ -120,7 +120,7 @@ define i32 @test_v3i32(<3 x i32> %a) nounwind {
 ; CHECK-NEXT:    addv s0, v0.4s
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.add.i32.v3i32(<3 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.add.v3i32(<3 x i32> %a)
   ret i32 %b
 }
 
@@ -131,7 +131,7 @@ define i1 @test_v4i1(<4 x i1> %a) nounwind {
 ; CHECK-NEXT:    fmov w8, s0
 ; CHECK-NEXT:    and w0, w8, #0x1
 ; CHECK-NEXT:    ret
-  %b = call i1 @llvm.experimental.vector.reduce.add.i1.v4i1(<4 x i1> %a)
+  %b = call i1 @llvm.experimental.vector.reduce.add.v4i1(<4 x i1> %a)
   ret i1 %b
 }
 
@@ -141,7 +141,7 @@ define i24 @test_v4i24(<4 x i24> %a) nounwind {
 ; CHECK-NEXT:    addv s0, v0.4s
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i24 @llvm.experimental.vector.reduce.add.i24.v4i24(<4 x i24> %a)
+  %b = call i24 @llvm.experimental.vector.reduce.add.v4i24(<4 x i24> %a)
   ret i24 %b
 }
 
@@ -151,7 +151,7 @@ define i128 @test_v2i128(<2 x i128> %a) nounwind {
 ; CHECK-NEXT:    adds x0, x0, x2
 ; CHECK-NEXT:    adcs x1, x1, x3
 ; CHECK-NEXT:    ret
-  %b = call i128 @llvm.experimental.vector.reduce.add.i128.v2i128(<2 x i128> %a)
+  %b = call i128 @llvm.experimental.vector.reduce.add.v2i128(<2 x i128> %a)
   ret i128 %b
 }
 
@@ -164,6 +164,6 @@ define i32 @test_v16i32(<16 x i32> %a) nounwind {
 ; CHECK-NEXT:    addv s0, v0.4s
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> %a)
   ret i32 %b
 }

test/CodeGen/AArch64/vecreduce-and-legalization.ll

@@ -1,28 +1,28 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s --check-prefix=CHECK
 
-declare i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> %a)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v1i8(<1 x i8> %a)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v1i16(<1 x i16> %a)
-declare i24 @llvm.experimental.vector.reduce.and.i24.v1i24(<1 x i24> %a)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v1i32(<1 x i32> %a)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> %a)
-declare i128 @llvm.experimental.vector.reduce.and.i128.v1i128(<1 x i128> %a)
-
-declare i8 @llvm.experimental.vector.reduce.and.i8.v3i8(<3 x i8> %a)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v9i8(<9 x i8> %a)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v3i32(<3 x i32> %a)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> %a)
-declare i24 @llvm.experimental.vector.reduce.and.i24.v4i24(<4 x i24> %a)
-declare i128 @llvm.experimental.vector.reduce.and.i128.v2i128(<2 x i128> %a)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> %a)
+declare i8 @llvm.experimental.vector.reduce.and.v1i8(<1 x i8> %a)
+declare i16 @llvm.experimental.vector.reduce.and.v1i16(<1 x i16> %a)
+declare i24 @llvm.experimental.vector.reduce.and.v1i24(<1 x i24> %a)
+declare i32 @llvm.experimental.vector.reduce.and.v1i32(<1 x i32> %a)
+declare i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> %a)
+declare i128 @llvm.experimental.vector.reduce.and.v1i128(<1 x i128> %a)
+
+declare i8 @llvm.experimental.vector.reduce.and.v3i8(<3 x i8> %a)
+declare i8 @llvm.experimental.vector.reduce.and.v9i8(<9 x i8> %a)
+declare i32 @llvm.experimental.vector.reduce.and.v3i32(<3 x i32> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> %a)
+declare i24 @llvm.experimental.vector.reduce.and.v4i24(<4 x i24> %a)
+declare i128 @llvm.experimental.vector.reduce.and.v2i128(<2 x i128> %a)
+declare i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> %a)
 
 define i1 @test_v1i1(<1 x i1> %a) nounwind {
 ; CHECK-LABEL: test_v1i1:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    and w0, w0, #0x1
 ; CHECK-NEXT:    ret
-  %b = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> %a)
+  %b = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> %a)
   ret i1 %b
 }
 
@@ -32,7 +32,7 @@ define i8 @test_v1i8(<1 x i8> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    umov w0, v0.b[0]
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.and.i8.v1i8(<1 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.and.v1i8(<1 x i8> %a)
   ret i8 %b
 }
 
@@ -42,7 +42,7 @@ define i16 @test_v1i16(<1 x i16> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    umov w0, v0.h[0]
 ; CHECK-NEXT:    ret
-  %b = call i16 @llvm.experimental.vector.reduce.and.i16.v1i16(<1 x i16> %a)
+  %b = call i16 @llvm.experimental.vector.reduce.and.v1i16(<1 x i16> %a)
   ret i16 %b
 }
 
@@ -50,7 +50,7 @@ define i24 @test_v1i24(<1 x i24> %a) nounwind {
 ; CHECK-LABEL: test_v1i24:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call i24 @llvm.experimental.vector.reduce.and.i24.v1i24(<1 x i24> %a)
+  %b = call i24 @llvm.experimental.vector.reduce.and.v1i24(<1 x i24> %a)
   ret i24 %b
 }
 
@@ -60,7 +60,7 @@ define i32 @test_v1i32(<1 x i32> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.and.i32.v1i32(<1 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.and.v1i32(<1 x i32> %a)
   ret i32 %b
 }
 
@@ -70,7 +70,7 @@ define i64 @test_v1i64(<1 x i64> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    fmov x0, d0
 ; CHECK-NEXT:    ret
-  %b = call i64 @llvm.experimental.vector.reduce.and.i64.v1i64(<1 x i64> %a)
+  %b = call i64 @llvm.experimental.vector.reduce.and.v1i64(<1 x i64> %a)
   ret i64 %b
 }
 
@@ -78,7 +78,7 @@ define i128 @test_v1i128(<1 x i128> %a) nounwind {
 ; CHECK-LABEL: test_v1i128:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call i128 @llvm.experimental.vector.reduce.and.i128.v1i128(<1 x i128> %a)
+  %b = call i128 @llvm.experimental.vector.reduce.and.v1i128(<1 x i128> %a)
   ret i128 %b
 }
 
@@ -89,7 +89,7 @@ define i8 @test_v3i8(<3 x i8> %a) nounwind {
 ; CHECK-NEXT:    and w8, w8, w2
 ; CHECK-NEXT:    and w0, w8, #0xff
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.and.i8.v3i8(<3 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.and.v3i8(<3 x i8> %a)
   ret i8 %b
 }
 
@@ -122,7 +122,7 @@ define i8 @test_v9i8(<9 x i8> %a) nounwind {
 ; CHECK-NEXT:    umov w9, v0.b[7]
 ; CHECK-NEXT:    and w0, w8, w9
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.and.i8.v9i8(<9 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.and.v9i8(<9 x i8> %a)
   ret i8 %b
 }
 
@@ -137,7 +137,7 @@ define i32 @test_v3i32(<3 x i32> %a) nounwind {
 ; CHECK-NEXT:    fmov w9, s0
 ; CHECK-NEXT:    and w0, w9, w8
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.and.i32.v3i32(<3 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.and.v3i32(<3 x i32> %a)
   ret i32 %b
 }
 
@@ -154,7 +154,7 @@ define i1 @test_v4i1(<4 x i1> %a) nounwind {
 ; CHECK-NEXT:    and w8, w9, w8
 ; CHECK-NEXT:    and w0, w8, #0x1
 ; CHECK-NEXT:    ret
-  %b = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> %a)
+  %b = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> %a)
   ret i1 %b
 }
 
@@ -167,7 +167,7 @@ define i24 @test_v4i24(<4 x i24> %a) nounwind {
 ; CHECK-NEXT:    fmov w9, s0
 ; CHECK-NEXT:    and w0, w9, w8
 ; CHECK-NEXT:    ret
-  %b = call i24 @llvm.experimental.vector.reduce.and.i24.v4i24(<4 x i24> %a)
+  %b = call i24 @llvm.experimental.vector.reduce.and.v4i24(<4 x i24> %a)
   ret i24 %b
 }
 
@@ -177,7 +177,7 @@ define i128 @test_v2i128(<2 x i128> %a) nounwind {
 ; CHECK-NEXT:    and x0, x0, x2
 ; CHECK-NEXT:    and x1, x1, x3
 ; CHECK-NEXT:    ret
-  %b = call i128 @llvm.experimental.vector.reduce.and.i128.v2i128(<2 x i128> %a)
+  %b = call i128 @llvm.experimental.vector.reduce.and.v2i128(<2 x i128> %a)
   ret i128 %b
 }
 
@@ -193,6 +193,6 @@ define i32 @test_v16i32(<16 x i32> %a) nounwind {
 ; CHECK-NEXT:    fmov w9, s0
 ; CHECK-NEXT:    and w0, w9, w8
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> %a)
   ret i32 %b
 }

test/CodeGen/AArch64/vecreduce-bool.ll

@@ -1,19 +1,19 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s --check-prefix=CHECK
 
-declare i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> %a)
 
-declare i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> %a)
-declare i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> %a)
+declare i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> %a)
 
 define i32 @reduce_and_v1(<1 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-LABEL: reduce_and_v1:
@@ -24,7 +24,7 @@ define i32 @reduce_and_v1(<1 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, lt
 ; CHECK-NEXT:    ret
   %x = icmp slt <1 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.and.i1.v1i1(<1 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.and.v1i1(<1 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -41,7 +41,7 @@ define i32 @reduce_and_v2(<2 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <2 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.and.i1.v2i1(<2 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.and.v2i1(<2 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -58,7 +58,7 @@ define i32 @reduce_and_v4(<4 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <4 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.and.i1.v4i1(<4 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.and.v4i1(<4 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -73,7 +73,7 @@ define i32 @reduce_and_v8(<8 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <8 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.and.i1.v8i1(<8 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.and.v8i1(<8 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -88,7 +88,7 @@ define i32 @reduce_and_v16(<16 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <16 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.and.i1.v16i1(<16 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.and.v16i1(<16 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -105,7 +105,7 @@ define i32 @reduce_and_v32(<32 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <32 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.and.i1.v32i1(<32 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.and.v32i1(<32 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -119,7 +119,7 @@ define i32 @reduce_or_v1(<1 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, lt
 ; CHECK-NEXT:    ret
   %x = icmp slt <1 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.or.i1.v1i1(<1 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.or.v1i1(<1 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -136,7 +136,7 @@ define i32 @reduce_or_v2(<2 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <2 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.or.i1.v2i1(<2 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.or.v2i1(<2 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -153,7 +153,7 @@ define i32 @reduce_or_v4(<4 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <4 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.or.i1.v4i1(<4 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.or.v4i1(<4 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -168,7 +168,7 @@ define i32 @reduce_or_v8(<8 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <8 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.or.i1.v8i1(<8 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.or.v8i1(<8 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -183,7 +183,7 @@ define i32 @reduce_or_v16(<16 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <16 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.or.i1.v16i1(<16 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.or.v16i1(<16 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }
@@ -200,7 +200,7 @@ define i32 @reduce_or_v32(<32 x i8> %a0, i32 %a1, i32 %a2) nounwind {
 ; CHECK-NEXT:    csel w0, w0, w1, ne
 ; CHECK-NEXT:    ret
   %x = icmp slt <32 x i8> %a0, zeroinitializer
-  %y = call i1 @llvm.experimental.vector.reduce.or.i1.v32i1(<32 x i1> %x)
+  %y = call i1 @llvm.experimental.vector.reduce.or.v32i1(<32 x i1> %x)
   %z = select i1 %y, i32 %a1, i32 %a2
   ret i32 %z
 }

test/CodeGen/AArch64/vecreduce-fmax-legalization.ll

@@ -1,20 +1,20 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s --check-prefix=CHECK
 
-declare half @llvm.experimental.vector.reduce.fmax.f16.v1f16(<1 x half> %a)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v1f32(<1 x float> %a)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v1f64(<1 x double> %a)
-declare fp128 @llvm.experimental.vector.reduce.fmax.f128.v1f128(<1 x fp128> %a)
+declare half @llvm.experimental.vector.reduce.fmax.v1f16(<1 x half> %a)
+declare float @llvm.experimental.vector.reduce.fmax.v1f32(<1 x float> %a)
+declare double @llvm.experimental.vector.reduce.fmax.v1f64(<1 x double> %a)
+declare fp128 @llvm.experimental.vector.reduce.fmax.v1f128(<1 x fp128> %a)
 
-declare float @llvm.experimental.vector.reduce.fmax.f32.v3f32(<3 x float> %a)
-declare fp128 @llvm.experimental.vector.reduce.fmax.f128.v2f128(<2 x fp128> %a)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v16f32(<16 x float> %a)
+declare float @llvm.experimental.vector.reduce.fmax.v3f32(<3 x float> %a)
+declare fp128 @llvm.experimental.vector.reduce.fmax.v2f128(<2 x fp128> %a)
+declare float @llvm.experimental.vector.reduce.fmax.v16f32(<16 x float> %a)
 
 define half @test_v1f16(<1 x half> %a) nounwind {
 ; CHECK-LABEL: test_v1f16:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call nnan half @llvm.experimental.vector.reduce.fmax.f16.v1f16(<1 x half> %a)
+  %b = call nnan half @llvm.experimental.vector.reduce.fmax.v1f16(<1 x half> %a)
   ret half %b
 }
 
@@ -24,7 +24,7 @@ define float @test_v1f32(<1 x float> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    // kill: def $s0 killed $s0 killed $q0
 ; CHECK-NEXT:    ret
-  %b = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v1f32(<1 x float> %a)
+  %b = call nnan float @llvm.experimental.vector.reduce.fmax.v1f32(<1 x float> %a)
   ret float %b
 }
 
@@ -32,7 +32,7 @@ define double @test_v1f64(<1 x double> %a) nounwind {
 ; CHECK-LABEL: test_v1f64:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call nnan double @llvm.experimental.vector.reduce.fmax.f64.v1f64(<1 x double> %a)
+  %b = call nnan double @llvm.experimental.vector.reduce.fmax.v1f64(<1 x double> %a)
   ret double %b
 }
 
@@ -40,7 +40,7 @@ define fp128 @test_v1f128(<1 x fp128> %a) nounwind {
 ; CHECK-LABEL: test_v1f128:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call nnan fp128 @llvm.experimental.vector.reduce.fmax.f128.v1f128(<1 x fp128> %a)
+  %b = call nnan fp128 @llvm.experimental.vector.reduce.fmax.v1f128(<1 x fp128> %a)
   ret fp128 %b
 }
 
@@ -52,7 +52,7 @@ define float @test_v3f32(<3 x float> %a) nounwind {
 ; CHECK-NEXT:    mov v0.s[3], v1.s[0]
 ; CHECK-NEXT:    fmaxnmv s0, v0.4s
 ; CHECK-NEXT:    ret
-  %b = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v3f32(<3 x float> %a)
+  %b = call nnan float @llvm.experimental.vector.reduce.fmax.v3f32(<3 x float> %a)
   ret float %b
 }
 
@@ -60,7 +60,7 @@ define fp128 @test_v2f128(<2 x fp128> %a) nounwind {
 ; CHECK-LABEL: test_v2f128:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    b fmaxl
-  %b = call nnan fp128 @llvm.experimental.vector.reduce.fmax.f128.v2f128(<2 x fp128> %a)
+  %b = call nnan fp128 @llvm.experimental.vector.reduce.fmax.v2f128(<2 x fp128> %a)
   ret fp128 %b
 }
 
@@ -72,6 +72,6 @@ define float @test_v16f32(<16 x float> %a) nounwind {
 ; CHECK-NEXT:    fmaxnm v0.4s, v0.4s, v1.4s
 ; CHECK-NEXT:    fmaxnmv s0, v0.4s
 ; CHECK-NEXT:    ret
-  %b = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v16f32(<16 x float> %a)
+  %b = call nnan float @llvm.experimental.vector.reduce.fmax.v16f32(<16 x float> %a)
   ret float %b
 }

test/CodeGen/AArch64/vecreduce-propagate-sd-flags.ll

@@ -24,8 +24,8 @@ entry:
  %1 = insertelement <4 x double> %0, double 1.0, i32 1
  %2 = insertelement <4 x double> %1, double 1.0, i32 2
  %3 = insertelement <4 x double> %2, double 1.0, i32 3
- %4 = call nnan reassoc double @llvm.experimental.vector.reduce.fmax.f64.v4f64(<4 x double> %3)
+ %4 = call nnan reassoc double @llvm.experimental.vector.reduce.fmax.v4f64(<4 x double> %3)
  ret double %4
 }
 
-declare double @llvm.experimental.vector.reduce.fmax.f64.v4f64(<4 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v4f64(<4 x double>)

test/CodeGen/AArch64/vecreduce-umax-legalization.ll

@@ -1,28 +1,28 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s --check-prefix=CHECK
 
-declare i1 @llvm.experimental.vector.reduce.umax.i1.v1i1(<1 x i1> %a)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v1i8(<1 x i8> %a)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v1i16(<1 x i16> %a)
-declare i24 @llvm.experimental.vector.reduce.umax.i24.v1i24(<1 x i24> %a)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v1i32(<1 x i32> %a)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> %a)
-declare i128 @llvm.experimental.vector.reduce.umax.i128.v1i128(<1 x i128> %a)
-
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v3i8(<3 x i8> %a)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v9i8(<9 x i8> %a)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v3i32(<3 x i32> %a)
-declare i1 @llvm.experimental.vector.reduce.umax.i1.v4i1(<4 x i1> %a)
-declare i24 @llvm.experimental.vector.reduce.umax.i24.v4i24(<4 x i24> %a)
-declare i128 @llvm.experimental.vector.reduce.umax.i128.v2i128(<2 x i128> %a)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> %a)
+declare i1 @llvm.experimental.vector.reduce.umax.v1i1(<1 x i1> %a)
+declare i8 @llvm.experimental.vector.reduce.umax.v1i8(<1 x i8> %a)
+declare i16 @llvm.experimental.vector.reduce.umax.v1i16(<1 x i16> %a)
+declare i24 @llvm.experimental.vector.reduce.umax.v1i24(<1 x i24> %a)
+declare i32 @llvm.experimental.vector.reduce.umax.v1i32(<1 x i32> %a)
+declare i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> %a)
+declare i128 @llvm.experimental.vector.reduce.umax.v1i128(<1 x i128> %a)
+
+declare i8 @llvm.experimental.vector.reduce.umax.v3i8(<3 x i8> %a)
+declare i8 @llvm.experimental.vector.reduce.umax.v9i8(<9 x i8> %a)
+declare i32 @llvm.experimental.vector.reduce.umax.v3i32(<3 x i32> %a)
+declare i1 @llvm.experimental.vector.reduce.umax.v4i1(<4 x i1> %a)
+declare i24 @llvm.experimental.vector.reduce.umax.v4i24(<4 x i24> %a)
+declare i128 @llvm.experimental.vector.reduce.umax.v2i128(<2 x i128> %a)
+declare i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> %a)
 
 define i1 @test_v1i1(<1 x i1> %a) nounwind {
 ; CHECK-LABEL: test_v1i1:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    and w0, w0, #0x1
 ; CHECK-NEXT:    ret
-  %b = call i1 @llvm.experimental.vector.reduce.umax.i1.v1i1(<1 x i1> %a)
+  %b = call i1 @llvm.experimental.vector.reduce.umax.v1i1(<1 x i1> %a)
   ret i1 %b
 }
 
@@ -32,7 +32,7 @@ define i8 @test_v1i8(<1 x i8> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    umov w0, v0.b[0]
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.umax.i8.v1i8(<1 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.umax.v1i8(<1 x i8> %a)
   ret i8 %b
 }
 
@@ -42,7 +42,7 @@ define i16 @test_v1i16(<1 x i16> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    umov w0, v0.h[0]
 ; CHECK-NEXT:    ret
-  %b = call i16 @llvm.experimental.vector.reduce.umax.i16.v1i16(<1 x i16> %a)
+  %b = call i16 @llvm.experimental.vector.reduce.umax.v1i16(<1 x i16> %a)
   ret i16 %b
 }
 
@@ -50,7 +50,7 @@ define i24 @test_v1i24(<1 x i24> %a) nounwind {
 ; CHECK-LABEL: test_v1i24:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call i24 @llvm.experimental.vector.reduce.umax.i24.v1i24(<1 x i24> %a)
+  %b = call i24 @llvm.experimental.vector.reduce.umax.v1i24(<1 x i24> %a)
   ret i24 %b
 }
 
@@ -60,7 +60,7 @@ define i32 @test_v1i32(<1 x i32> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.umax.i32.v1i32(<1 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.umax.v1i32(<1 x i32> %a)
   ret i32 %b
 }
 
@@ -70,7 +70,7 @@ define i64 @test_v1i64(<1 x i64> %a) nounwind {
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
 ; CHECK-NEXT:    fmov x0, d0
 ; CHECK-NEXT:    ret
-  %b = call i64 @llvm.experimental.vector.reduce.umax.i64.v1i64(<1 x i64> %a)
+  %b = call i64 @llvm.experimental.vector.reduce.umax.v1i64(<1 x i64> %a)
   ret i64 %b
 }
 
@@ -78,7 +78,7 @@ define i128 @test_v1i128(<1 x i128> %a) nounwind {
 ; CHECK-LABEL: test_v1i128:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %b = call i128 @llvm.experimental.vector.reduce.umax.i128.v1i128(<1 x i128> %a)
+  %b = call i128 @llvm.experimental.vector.reduce.umax.v1i128(<1 x i128> %a)
   ret i128 %b
 }
 
@@ -93,7 +93,7 @@ define i8 @test_v3i8(<3 x i8> %a) nounwind {
 ; CHECK-NEXT:    umaxv h0, v0.4h
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.umax.i8.v3i8(<3 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.umax.v3i8(<3 x i8> %a)
   ret i8 %b
 }
 
@@ -110,7 +110,7 @@ define i8 @test_v9i8(<9 x i8> %a) nounwind {
 ; CHECK-NEXT:    umaxv b0, v0.16b
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i8 @llvm.experimental.vector.reduce.umax.i8.v9i8(<9 x i8> %a)
+  %b = call i8 @llvm.experimental.vector.reduce.umax.v9i8(<9 x i8> %a)
   ret i8 %b
 }
 
@@ -121,7 +121,7 @@ define i32 @test_v3i32(<3 x i32> %a) nounwind {
 ; CHECK-NEXT:    umaxv s0, v0.4s
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.umax.i32.v3i32(<3 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.umax.v3i32(<3 x i32> %a)
   ret i32 %b
 }
 
@@ -134,7 +134,7 @@ define i1 @test_v4i1(<4 x i1> %a) nounwind {
 ; CHECK-NEXT:    fmov w8, s0
 ; CHECK-NEXT:    and w0, w8, #0x1
 ; CHECK-NEXT:    ret
-  %b = call i1 @llvm.experimental.vector.reduce.umax.i1.v4i1(<4 x i1> %a)
+  %b = call i1 @llvm.experimental.vector.reduce.umax.v4i1(<4 x i1> %a)
   ret i1 %b
 }
 
@@ -145,7 +145,7 @@ define i24 @test_v4i24(<4 x i24> %a) nounwind {
 ; CHECK-NEXT:    umaxv s0, v0.4s
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i24 @llvm.experimental.vector.reduce.umax.i24.v4i24(<4 x i24> %a)
+  %b = call i24 @llvm.experimental.vector.reduce.umax.v4i24(<4 x i24> %a)
   ret i24 %b
 }
 
@@ -159,7 +159,7 @@ define i128 @test_v2i128(<2 x i128> %a) nounwind {
 ; CHECK-NEXT:    csel x0, x8, x9, eq
 ; CHECK-NEXT:    csel x1, x1, x3, hi
 ; CHECK-NEXT:    ret
-  %b = call i128 @llvm.experimental.vector.reduce.umax.i128.v2i128(<2 x i128> %a)
+  %b = call i128 @llvm.experimental.vector.reduce.umax.v2i128(<2 x i128> %a)
   ret i128 %b
 }
 
@@ -172,6 +172,6 @@ define i32 @test_v16i32(<16 x i32> %a) nounwind {
 ; CHECK-NEXT:    umaxv s0, v0.4s
 ; CHECK-NEXT:    fmov w0, s0
 ; CHECK-NEXT:    ret
-  %b = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> %a)
+  %b = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> %a)
   ret i32 %b
 }

test/CodeGen/Generic/expand-experimental-reductions.ll

@@ -1,22 +1,22 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -expand-reductions -S | FileCheck %s
 ; Tests without a target which should expand all reductions
-declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64>)
 
 declare float @llvm.experimental.vector.reduce.v2.fadd.f32.v4f32(float, <4 x float>)
 declare float @llvm.experimental.vector.reduce.v2.fmul.f32.v4f32(float, <4 x float>)
 
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64>)
 
-declare double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double>)
 
 
 define i64 @add_i64(<2 x i64> %vec) {
@@ -28,7 +28,7 @@ define i64 @add_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -41,7 +41,7 @@ define i64 @mul_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -54,7 +54,7 @@ define i64 @and_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -67,7 +67,7 @@ define i64 @or_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -80,7 +80,7 @@ define i64 @xor_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -230,7 +230,7 @@ define i64 @smax_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -244,7 +244,7 @@ define i64 @smin_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -258,7 +258,7 @@ define i64 @umax_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -272,7 +272,7 @@ define i64 @umin_i64(<2 x i64> %vec) {
 ; CHECK-NEXT:    ret i64 [[TMP0]]
 ;
 entry:
-  %r = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> %vec)
+  %r = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> %vec)
   ret i64 %r
 }
 
@@ -286,7 +286,7 @@ define double @fmax_f64(<2 x double> %vec) {
 ; CHECK-NEXT:    ret double [[TMP0]]
 ;
 entry:
-  %r = call double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double> %vec)
+  %r = call double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double> %vec)
   ret double %r
 }
 
@@ -300,6 +300,6 @@ define double @fmin_f64(<2 x double> %vec) {
 ; CHECK-NEXT:    ret double [[TMP0]]
 ;
 entry:
-  %r = call double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double> %vec)
+  %r = call double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double> %vec)
   ret double %r
 }

test/CodeGen/X86/vector-reduce-add-widen.ll

@@ -32,7 +32,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512-NEXT:    vpaddq %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -74,7 +74,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -124,7 +124,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -187,7 +187,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -229,7 +229,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpaddd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -277,7 +277,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpaddd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -338,7 +338,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -410,7 +410,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -501,7 +501,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -549,7 +549,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -603,7 +603,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -667,7 +667,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -744,7 +744,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -832,7 +832,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -939,7 +939,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -981,7 +981,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1029,7 +1029,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1085,7 +1085,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1149,7 +1149,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -1238,7 +1238,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1337,7 +1337,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1453,32 +1453,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-add.ll

@@ -32,7 +32,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512-NEXT:    vpaddq %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -74,7 +74,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -124,7 +124,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -187,7 +187,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -216,7 +216,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpaddq %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -264,7 +264,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpaddd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -325,7 +325,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -397,7 +397,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -488,7 +488,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -520,7 +520,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -573,7 +573,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -637,7 +637,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -714,7 +714,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -802,7 +802,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -909,7 +909,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -949,7 +949,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1012,7 +1012,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1089,7 +1089,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1153,7 +1153,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -1242,7 +1242,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1341,7 +1341,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1457,32 +1457,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.add.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.add.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.add.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.add.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.add.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.add.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.add.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-and-widen.ll

@@ -24,7 +24,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovq %xmm0, %rax
 ; AVX-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -66,7 +66,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -114,7 +114,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -171,7 +171,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -193,7 +193,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -215,7 +215,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -265,7 +265,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -321,7 +321,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -386,7 +386,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -411,7 +411,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -436,7 +436,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -465,7 +465,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -528,7 +528,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -597,7 +597,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -675,7 +675,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -709,7 +709,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -747,7 +747,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -791,7 +791,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -841,7 +841,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -930,7 +930,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1027,7 +1027,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1137,32 +1137,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-and.ll

@@ -24,7 +24,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovq %xmm0, %rax
 ; AVX-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -66,7 +66,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -114,7 +114,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -171,7 +171,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -193,7 +193,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -215,7 +215,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -265,7 +265,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -321,7 +321,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -386,7 +386,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -410,7 +410,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -434,7 +434,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -463,7 +463,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -526,7 +526,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -595,7 +595,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -673,7 +673,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -705,7 +705,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -739,7 +739,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -781,7 +781,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -831,7 +831,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -920,7 +920,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1017,7 +1017,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1127,32 +1127,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.and.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.and.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.and.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.and.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.and.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.and.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.and.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.and.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.and.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.and.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.and.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.and.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-fmax-nnan.ll

@@ -35,7 +35,7 @@ define float @test_v2f32(<2 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v2f32(<2 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.v2f32(<2 x float> %a0)
   ret float %1
 }
 
@@ -74,7 +74,7 @@ define float @test_v4f32(<4 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float> %a0)
   ret float %1
 }
 
@@ -121,7 +121,7 @@ define float @test_v8f32(<8 x float> %a0) {
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v8f32(<8 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.v8f32(<8 x float> %a0)
   ret float %1
 }
 
@@ -175,7 +175,7 @@ define float @test_v16f32(<16 x float> %a0) {
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v16f32(<16 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmax.v16f32(<16 x float> %a0)
   ret float %1
 }
 
@@ -202,7 +202,7 @@ define double @test_v2f64(<2 x double> %a0) {
 ; AVX512-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double> %a0)
   ret double %1
 }
 
@@ -232,7 +232,7 @@ define double @test_v4f64(<4 x double> %a0) {
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.f64.v4f64(<4 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.v4f64(<4 x double> %a0)
   ret double %1
 }
 
@@ -267,7 +267,7 @@ define double @test_v8f64(<8 x double> %a0) {
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.f64.v8f64(<8 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.v8f64(<8 x double> %a0)
   ret double %1
 }
 
@@ -309,16 +309,16 @@ define double @test_v16f64(<16 x double> %a0) {
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.f64.v16f64(<16 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmax.v16f64(<16 x double> %a0)
   ret double %1
 }
 
-declare float @llvm.experimental.vector.reduce.fmax.f32.v2f32(<2 x float>)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float>)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v8f32(<8 x float>)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v16f32(<16 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v2f32(<2 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v8f32(<8 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v16f32(<16 x float>)
 
-declare double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double>)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v4f64(<4 x double>)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v8f64(<8 x double>)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v16f64(<16 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v4f64(<4 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v8f64(<8 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v16f64(<16 x double>)

test/CodeGen/X86/vector-reduce-fmax.ll

@@ -35,7 +35,7 @@ define float @test_v2f32(<2 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmax.f32.v2f32(<2 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmax.v2f32(<2 x float> %a0)
   ret float %1
 }
 
@@ -74,7 +74,7 @@ define float @test_v4f32(<4 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float> %a0)
   ret float %1
 }
 
@@ -121,7 +121,7 @@ define float @test_v8f32(<8 x float> %a0) {
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmax.f32.v8f32(<8 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmax.v8f32(<8 x float> %a0)
   ret float %1
 }
 
@@ -175,7 +175,7 @@ define float @test_v16f32(<16 x float> %a0) {
 ; AVX512-NEXT:    vmaxss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmax.f32.v16f32(<16 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmax.v16f32(<16 x float> %a0)
   ret float %1
 }
 
@@ -202,7 +202,7 @@ define double @test_v2f64(<2 x double> %a0) {
 ; AVX512-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double> %a0)
   ret double %1
 }
 
@@ -232,7 +232,7 @@ define double @test_v4f64(<4 x double> %a0) {
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmax.f64.v4f64(<4 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmax.v4f64(<4 x double> %a0)
   ret double %1
 }
 
@@ -267,7 +267,7 @@ define double @test_v8f64(<8 x double> %a0) {
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmax.f64.v8f64(<8 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmax.v8f64(<8 x double> %a0)
   ret double %1
 }
 
@@ -309,16 +309,16 @@ define double @test_v16f64(<16 x double> %a0) {
 ; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmax.f64.v16f64(<16 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmax.v16f64(<16 x double> %a0)
   ret double %1
 }
 
-declare float @llvm.experimental.vector.reduce.fmax.f32.v2f32(<2 x float>)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float>)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v8f32(<8 x float>)
-declare float @llvm.experimental.vector.reduce.fmax.f32.v16f32(<16 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v2f32(<2 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v4f32(<4 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v8f32(<8 x float>)
+declare float @llvm.experimental.vector.reduce.fmax.v16f32(<16 x float>)
 
-declare double @llvm.experimental.vector.reduce.fmax.f64.v2f64(<2 x double>)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v4f64(<4 x double>)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v8f64(<8 x double>)
-declare double @llvm.experimental.vector.reduce.fmax.f64.v16f64(<16 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v2f64(<2 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v4f64(<4 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v8f64(<8 x double>)
+declare double @llvm.experimental.vector.reduce.fmax.v16f64(<16 x double>)

test/CodeGen/X86/vector-reduce-fmin-nnan.ll

@@ -35,7 +35,7 @@ define float @test_v2f32(<2 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v2f32(<2 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.v2f32(<2 x float> %a0)
   ret float %1
 }
 
@@ -74,7 +74,7 @@ define float @test_v4f32(<4 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float> %a0)
   ret float %1
 }
 
@@ -121,7 +121,7 @@ define float @test_v8f32(<8 x float> %a0) {
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v8f32(<8 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.v8f32(<8 x float> %a0)
   ret float %1
 }
 
@@ -175,7 +175,7 @@ define float @test_v16f32(<16 x float> %a0) {
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v16f32(<16 x float> %a0)
+  %1 = call nnan float @llvm.experimental.vector.reduce.fmin.v16f32(<16 x float> %a0)
   ret float %1
 }
 
@@ -202,7 +202,7 @@ define double @test_v2f64(<2 x double> %a0) {
 ; AVX512-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double> %a0)
   ret double %1
 }
 
@@ -232,7 +232,7 @@ define double @test_v4f64(<4 x double> %a0) {
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.f64.v4f64(<4 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.v4f64(<4 x double> %a0)
   ret double %1
 }
 
@@ -267,7 +267,7 @@ define double @test_v8f64(<8 x double> %a0) {
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.f64.v8f64(<8 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.v8f64(<8 x double> %a0)
   ret double %1
 }
 
@@ -309,16 +309,16 @@ define double @test_v16f64(<16 x double> %a0) {
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.f64.v16f64(<16 x double> %a0)
+  %1 = call nnan double @llvm.experimental.vector.reduce.fmin.v16f64(<16 x double> %a0)
   ret double %1
 }
 
-declare float @llvm.experimental.vector.reduce.fmin.f32.v2f32(<2 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v8f32(<8 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v16f32(<16 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v2f32(<2 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v8f32(<8 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v16f32(<16 x float>)
 
-declare double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v4f64(<4 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v8f64(<8 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v16f64(<16 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v4f64(<4 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v8f64(<8 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v16f64(<16 x double>)

test/CodeGen/X86/vector-reduce-fmin.ll

@@ -35,7 +35,7 @@ define float @test_v2f32(<2 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmin.f32.v2f32(<2 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmin.v2f32(<2 x float> %a0)
   ret float %1
 }
 
@@ -74,7 +74,7 @@ define float @test_v4f32(<4 x float> %a0) {
 ; AVX512-NEXT:    vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float> %a0)
   ret float %1
 }
 
@@ -121,7 +121,7 @@ define float @test_v8f32(<8 x float> %a0) {
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmin.f32.v8f32(<8 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmin.v8f32(<8 x float> %a0)
   ret float %1
 }
 
@@ -175,7 +175,7 @@ define float @test_v16f32(<16 x float> %a0) {
 ; AVX512-NEXT:    vminss %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call float @llvm.experimental.vector.reduce.fmin.f32.v16f32(<16 x float> %a0)
+  %1 = call float @llvm.experimental.vector.reduce.fmin.v16f32(<16 x float> %a0)
   ret float %1
 }
 
@@ -202,7 +202,7 @@ define double @test_v2f64(<2 x double> %a0) {
 ; AVX512-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double> %a0)
   ret double %1
 }
 
@@ -232,7 +232,7 @@ define double @test_v4f64(<4 x double> %a0) {
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmin.f64.v4f64(<4 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmin.v4f64(<4 x double> %a0)
   ret double %1
 }
 
@@ -267,7 +267,7 @@ define double @test_v8f64(<8 x double> %a0) {
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmin.f64.v8f64(<8 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmin.v8f64(<8 x double> %a0)
   ret double %1
 }
 
@@ -309,16 +309,16 @@ define double @test_v16f64(<16 x double> %a0) {
 ; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call double @llvm.experimental.vector.reduce.fmin.f64.v16f64(<16 x double> %a0)
+  %1 = call double @llvm.experimental.vector.reduce.fmin.v16f64(<16 x double> %a0)
   ret double %1
 }
 
-declare float @llvm.experimental.vector.reduce.fmin.f32.v2f32(<2 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v8f32(<8 x float>)
-declare float @llvm.experimental.vector.reduce.fmin.f32.v16f32(<16 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v2f32(<2 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v4f32(<4 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v8f32(<8 x float>)
+declare float @llvm.experimental.vector.reduce.fmin.v16f32(<16 x float>)
 
-declare double @llvm.experimental.vector.reduce.fmin.f64.v2f64(<2 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v4f64(<4 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v8f64(<8 x double>)
-declare double @llvm.experimental.vector.reduce.fmin.f64.v16f64(<16 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v2f64(<2 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v4f64(<4 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v8f64(<8 x double>)
+declare double @llvm.experimental.vector.reduce.fmin.v16f64(<16 x double>)

test/CodeGen/X86/vector-reduce-mul-widen.ll

@@ -86,7 +86,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vpmullq %xmm1, %xmm0, %xmm0
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -233,7 +233,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -446,7 +446,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -767,7 +767,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -803,7 +803,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpmulld %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -849,7 +849,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpmulld %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -920,7 +920,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -1005,7 +1005,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1115,7 +1115,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1148,7 +1148,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1183,7 +1183,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1224,7 +1224,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1287,7 +1287,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1407,7 +1407,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1545,7 +1545,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1587,7 +1587,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1647,7 +1647,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1726,7 +1726,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1945,7 +1945,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -2220,7 +2220,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2599,7 +2599,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -3086,32 +3086,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-mul.ll

@@ -86,7 +86,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vpmullq %xmm1, %xmm0, %xmm0
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -233,7 +233,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -446,7 +446,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -767,7 +767,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512DQVL-NEXT:    vmovq %xmm0, %rax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -796,7 +796,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpmuludq %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -842,7 +842,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpmulld %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -913,7 +913,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -998,7 +998,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1108,7 +1108,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1140,7 +1140,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1190,7 +1190,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1231,7 +1231,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1294,7 +1294,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1414,7 +1414,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1552,7 +1552,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1592,7 +1592,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1642,7 +1642,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1696,7 +1696,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1909,7 +1909,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -2182,7 +2182,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2555,7 +2555,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -3032,32 +3032,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512DQVL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512DQVL-NEXT:    vzeroupper
 ; AVX512DQVL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.mul.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.mul.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.mul.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.mul.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.mul.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.mul.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.mul.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.mul.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-or-widen.ll

@@ -24,7 +24,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovq %xmm0, %rax
 ; AVX-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -66,7 +66,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -114,7 +114,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -171,7 +171,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -193,7 +193,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -215,7 +215,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -265,7 +265,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -321,7 +321,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -386,7 +386,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -411,7 +411,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -436,7 +436,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -465,7 +465,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -528,7 +528,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -597,7 +597,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -675,7 +675,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -709,7 +709,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -747,7 +747,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -791,7 +791,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -841,7 +841,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -930,7 +930,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1027,7 +1027,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1137,32 +1137,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-or.ll

@@ -24,7 +24,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovq %xmm0, %rax
 ; AVX-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -66,7 +66,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -114,7 +114,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -171,7 +171,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -193,7 +193,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -215,7 +215,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -265,7 +265,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -321,7 +321,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -386,7 +386,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -410,7 +410,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -434,7 +434,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -463,7 +463,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -526,7 +526,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -595,7 +595,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -673,7 +673,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -705,7 +705,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -739,7 +739,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -781,7 +781,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -831,7 +831,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -920,7 +920,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1017,7 +1017,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1127,32 +1127,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.or.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.or.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.or.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.or.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.or.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.or.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.or.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.or.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.or.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.or.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.or.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.or.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-smax-widen.ll

@@ -74,7 +74,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpmaxsq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -190,7 +190,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -370,7 +370,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -674,7 +674,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -714,7 +714,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -762,7 +762,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -834,7 +834,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -924,7 +924,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1047,7 +1047,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1080,7 +1080,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1115,7 +1115,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1159,7 +1159,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    xorl $32767, %eax # imm = 0x7FFF
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1223,7 +1223,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1297,7 +1297,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1388,7 +1388,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1434,7 +1434,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1490,7 +1490,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1558,7 +1558,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1628,7 +1628,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    xorb $127, %al
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -1724,7 +1724,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1838,7 +1838,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1985,32 +1985,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-smax.ll

@@ -74,7 +74,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpmaxsq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -190,7 +190,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -370,7 +370,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -674,7 +674,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -788,7 +788,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512VL-NEXT:    vpmaxsq %xmm0, %xmm1, %xmm0
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -836,7 +836,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpmaxsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -908,7 +908,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -998,7 +998,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1121,7 +1121,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1259,7 +1259,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1335,7 +1335,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1379,7 +1379,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    xorl $32767, %eax # imm = 0x7FFF
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1443,7 +1443,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1517,7 +1517,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1608,7 +1608,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1746,7 +1746,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512VL-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1822,7 +1822,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1908,7 +1908,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1978,7 +1978,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    xorb $127, %al
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -2074,7 +2074,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2188,7 +2188,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -2335,32 +2335,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smax.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-smin-widen.ll

@@ -74,7 +74,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpminsq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -189,7 +189,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -369,7 +369,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -673,7 +673,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -713,7 +713,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -761,7 +761,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -833,7 +833,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -923,7 +923,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1046,7 +1046,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1079,7 +1079,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1114,7 +1114,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1158,7 +1158,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    xorl $32768, %eax # imm = 0x8000
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1222,7 +1222,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1296,7 +1296,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1387,7 +1387,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1433,7 +1433,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1489,7 +1489,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1557,7 +1557,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1627,7 +1627,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    xorb $-128, %al
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -1723,7 +1723,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1837,7 +1837,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1984,32 +1984,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-smin.ll

@@ -74,7 +74,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpminsq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -189,7 +189,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -369,7 +369,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -673,7 +673,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -787,7 +787,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512VL-NEXT:    vpminsq %xmm0, %xmm1, %xmm0
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -835,7 +835,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpminsd %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -907,7 +907,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -997,7 +997,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1120,7 +1120,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1258,7 +1258,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1334,7 +1334,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1378,7 +1378,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    xorl $32768, %eax # imm = 0x8000
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1442,7 +1442,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1516,7 +1516,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1607,7 +1607,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1745,7 +1745,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512VL-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1821,7 +1821,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1907,7 +1907,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1977,7 +1977,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    xorb $-128, %al
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -2073,7 +2073,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2187,7 +2187,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -2334,32 +2334,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.smin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.smin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.smin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.smin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.smin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.smin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.smin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.smin.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-umax-widen.ll

@@ -77,7 +77,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpmaxuq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -203,7 +203,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -399,7 +399,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -731,7 +731,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -774,7 +774,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpmaxud %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -828,7 +828,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpmaxud %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -909,7 +909,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -1014,7 +1014,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1164,7 +1164,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1210,7 +1210,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1262,7 +1262,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1328,7 +1328,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512VL-NEXT:    notl %eax
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1419,7 +1419,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1524,7 +1524,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1651,7 +1651,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1693,7 +1693,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1741,7 +1741,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1797,7 +1797,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1866,7 +1866,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512VL-NEXT:    notb %al
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -1959,7 +1959,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2064,7 +2064,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -2187,32 +2187,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-umax.ll

@@ -77,7 +77,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpmaxuq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -203,7 +203,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -399,7 +399,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -731,7 +731,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -823,7 +823,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512VL-NEXT:    vpmaxuq %xmm0, %xmm1, %xmm0
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -877,7 +877,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpmaxud %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -958,7 +958,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -1063,7 +1063,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1213,7 +1213,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1303,7 +1303,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1377,7 +1377,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1443,7 +1443,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512VL-NEXT:    notl %eax
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1534,7 +1534,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1639,7 +1639,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1766,7 +1766,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1854,7 +1854,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512VL-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1941,7 +1941,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -2037,7 +2037,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -2106,7 +2106,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512VL-NEXT:    notb %al
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -2199,7 +2199,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2304,7 +2304,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -2427,32 +2427,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umax.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umax.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umax.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umax.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umax.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umax.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umax.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umax.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-umin-widen.ll

@@ -77,7 +77,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpminuq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -202,7 +202,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -398,7 +398,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -730,7 +730,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -773,7 +773,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512-NEXT:    vpminud %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -827,7 +827,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpminud %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -908,7 +908,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -1013,7 +1013,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1163,7 +1163,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1209,7 +1209,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1261,7 +1261,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1308,7 +1308,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1376,7 +1376,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1456,7 +1456,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1557,7 +1557,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1599,7 +1599,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1647,7 +1647,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1703,7 +1703,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1751,7 +1751,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -1819,7 +1819,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1897,7 +1897,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1992,32 +1992,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-umin.ll

@@ -77,7 +77,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX512VL-NEXT:    vpminuq %xmm1, %xmm0, %xmm0
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -202,7 +202,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -398,7 +398,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -730,7 +730,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512VL-NEXT:    vmovq %xmm0, %rax
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -822,7 +822,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX512VL-NEXT:    vpminuq %xmm0, %xmm1, %xmm0
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -876,7 +876,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX512-NEXT:    vpminud %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -957,7 +957,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -1062,7 +1062,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -1212,7 +1212,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -1302,7 +1302,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX512VL-NEXT:    vmovd %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -1376,7 +1376,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -1423,7 +1423,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -1491,7 +1491,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -1571,7 +1571,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -1672,7 +1672,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -1760,7 +1760,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX512VL-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512VL-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512VL-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -1847,7 +1847,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -1943,7 +1943,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -1991,7 +1991,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX512-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -2059,7 +2059,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -2137,7 +2137,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -2232,32 +2232,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.umin.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.umin.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.umin.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.umin.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.umin.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.umin.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.umin.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.umin.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-xor-widen.ll

@@ -24,7 +24,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX-NEXT:    vpxor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovq %xmm0, %rax
 ; AVX-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -66,7 +66,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -114,7 +114,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -171,7 +171,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -193,7 +193,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX-NEXT:    vpxor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -215,7 +215,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX-NEXT:    vpxor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -265,7 +265,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -321,7 +321,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -386,7 +386,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -411,7 +411,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -436,7 +436,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -465,7 +465,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -528,7 +528,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -597,7 +597,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -675,7 +675,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -709,7 +709,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -747,7 +747,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -791,7 +791,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -841,7 +841,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -930,7 +930,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1027,7 +1027,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1137,32 +1137,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8>)

test/CodeGen/X86/vector-reduce-xor.ll

@@ -24,7 +24,7 @@ define i64 @test_v2i64(<2 x i64> %a0) {
 ; AVX-NEXT:    vpxor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovq %xmm0, %rax
 ; AVX-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64> %a0)
   ret i64 %1
 }
 
@@ -66,7 +66,7 @@ define i64 @test_v4i64(<4 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64> %a0)
   ret i64 %1
 }
 
@@ -114,7 +114,7 @@ define i64 @test_v8i64(<8 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64> %a0)
   ret i64 %1
 }
 
@@ -171,7 +171,7 @@ define i64 @test_v16i64(<16 x i64> %a0) {
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64> %a0)
+  %1 = call i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64> %a0)
   ret i64 %1
 }
 
@@ -193,7 +193,7 @@ define i32 @test_v2i32(<2 x i32> %a0) {
 ; AVX-NEXT:    vpxor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32> %a0)
   ret i32 %1
 }
 
@@ -215,7 +215,7 @@ define i32 @test_v4i32(<4 x i32> %a0) {
 ; AVX-NEXT:    vpxor %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> %a0)
   ret i32 %1
 }
 
@@ -265,7 +265,7 @@ define i32 @test_v8i32(<8 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32> %a0)
   ret i32 %1
 }
 
@@ -321,7 +321,7 @@ define i32 @test_v16i32(<16 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32> %a0)
   ret i32 %1
 }
 
@@ -386,7 +386,7 @@ define i32 @test_v32i32(<32 x i32> %a0) {
 ; AVX512-NEXT:    vmovd %xmm0, %eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32> %a0)
+  %1 = call i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32> %a0)
   ret i32 %1
 }
 
@@ -410,7 +410,7 @@ define i16 @test_v2i16(<2 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16> %a0)
   ret i16 %1
 }
 
@@ -434,7 +434,7 @@ define i16 @test_v4i16(<4 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16> %a0)
   ret i16 %1
 }
 
@@ -463,7 +463,7 @@ define i16 @test_v8i16(<8 x i16> %a0) {
 ; AVX-NEXT:    vmovd %xmm0, %eax
 ; AVX-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16> %a0)
   ret i16 %1
 }
 
@@ -526,7 +526,7 @@ define i16 @test_v16i16(<16 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16> %a0)
   ret i16 %1
 }
 
@@ -595,7 +595,7 @@ define i16 @test_v32i16(<32 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16> %a0)
   ret i16 %1
 }
 
@@ -673,7 +673,7 @@ define i16 @test_v64i16(<64 x i16> %a0) {
 ; AVX512-NEXT:    # kill: def $ax killed $ax killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16> %a0)
+  %1 = call i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16> %a0)
   ret i16 %1
 }
 
@@ -705,7 +705,7 @@ define i8 @test_v2i8(<2 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8> %a0)
   ret i8 %1
 }
 
@@ -739,7 +739,7 @@ define i8 @test_v4i8(<4 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8> %a0)
   ret i8 %1
 }
 
@@ -781,7 +781,7 @@ define i8 @test_v8i8(<8 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8> %a0)
   ret i8 %1
 }
 
@@ -831,7 +831,7 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8> %a0)
   ret i8 %1
 }
 
@@ -920,7 +920,7 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8> %a0)
   ret i8 %1
 }
 
@@ -1017,7 +1017,7 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8> %a0)
   ret i8 %1
 }
 
@@ -1127,32 +1127,32 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX512-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX512-NEXT:    vzeroupper
 ; AVX512-NEXT:    retq
-  %1 = call i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8> %a0)
+  %1 = call i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8> %a0)
   ret i8 %1
 }
 
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v2i64(<2 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v4i64(<4 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v8i64(<8 x i64>)
-declare i64 @llvm.experimental.vector.reduce.xor.i64.v16i64(<16 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v2i64(<2 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v4i64(<4 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v8i64(<8 x i64>)
+declare i64 @llvm.experimental.vector.reduce.xor.v16i64(<16 x i64>)
 
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v2i32(<2 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v4i32(<4 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v8i32(<8 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v16i32(<16 x i32>)
-declare i32 @llvm.experimental.vector.reduce.xor.i32.v32i32(<32 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v2i32(<2 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v8i32(<8 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v16i32(<16 x i32>)
+declare i32 @llvm.experimental.vector.reduce.xor.v32i32(<32 x i32>)
 
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v2i16(<2 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v4i16(<4 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v8i16(<8 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v16i16(<16 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v32i16(<32 x i16>)
-declare i16 @llvm.experimental.vector.reduce.xor.i16.v64i16(<64 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v2i16(<2 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v4i16(<4 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v8i16(<8 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v16i16(<16 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v32i16(<32 x i16>)
+declare i16 @llvm.experimental.vector.reduce.xor.v64i16(<64 x i16>)
 
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v2i8(<2 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v4i8(<4 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v8i8(<8 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v16i8(<16 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v32i8(<32 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v64i8(<64 x i8>)
-declare i8 @llvm.experimental.vector.reduce.xor.i8.v128i8(<128 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v2i8(<2 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v4i8(<4 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v8i8(<8 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v16i8(<16 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v32i8(<32 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v64i8(<64 x i8>)
+declare i8 @llvm.experimental.vector.reduce.xor.v128i8(<128 x i8>)

test/Transforms/LoopVectorize/AArch64/reduction-small-size.ll

@@ -20,7 +20,7 @@ target triple = "aarch64--linux-gnu"
 ; CHECK:   add <16 x i8>
 ;
 ; CHECK: middle.block:
-; CHECK:   [[Rdx:%[a-zA-Z0-9.]+]] = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>
+; CHECK:   [[Rdx:%[a-zA-Z0-9.]+]] = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8>
 ; CHECK:   zext i8 [[Rdx]] to i32
 ;
 define i8 @reduction_i8(i8* nocapture readonly %a, i8* nocapture readonly %b, i32 %n) {
@@ -75,7 +75,7 @@ for.body:
 ; CHECK:   add <8 x i16>
 ;
 ; CHECK: middle.block:
-; CHECK:   [[Rdx:%[a-zA-Z0-9.]+]] = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>
+; CHECK:   [[Rdx:%[a-zA-Z0-9.]+]] = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>
 ; CHECK:   zext i16 [[Rdx]] to i32
 ;
 define i16 @reduction_i16_1(i16* nocapture readonly %a, i16* nocapture readonly %b, i32 %n) {
@@ -132,7 +132,7 @@ for.body:
 ; CHECK:   add <8 x i16>
 ;
 ; CHECK: middle.block:
-; CHECK:   [[Rdx:%[a-zA-Z0-9.]+]] = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>
+; CHECK:   [[Rdx:%[a-zA-Z0-9.]+]] = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16>
 ; CHECK:   zext i16 [[Rdx]] to i32
 ;
 define i16 @reduction_i16_2(i8* nocapture readonly %a, i8* nocapture readonly %b, i32 %n) {

test/Transforms/SLPVectorizer/AArch64/gather-cost.ll

@@ -23,7 +23,7 @@ define internal i32 @gather_multiple_use(i32 %a, i32 %b, i32 %c, i32 %d) {
 ; CHECK-NEXT:    [[TMP7:%.*]] = mul nuw <4 x i32> [[TMP6]], <i32 65535, i32 65535, i32 65535, i32 65535>
 ; CHECK-NEXT:    [[TMP8:%.*]] = add <4 x i32> [[TMP7]], [[TMP4]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = xor <4 x i32> [[TMP8]], [[TMP7]]
-; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> [[TMP9]])
+; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP9]])
 ; CHECK-NEXT:    ret i32 [[TMP10]]
 ;
   %tmp00 = lshr i32 %a, 15

test/Transforms/SLPVectorizer/AArch64/gather-root.ll

@@ -24,7 +24,7 @@ define void @PR28330(i32 %n) {
 ; DEFAULT-NEXT:    [[P28:%.*]] = add i32 [[P26]], undef
 ; DEFAULT-NEXT:    [[P30:%.*]] = add i32 [[P28]], undef
 ; DEFAULT-NEXT:    [[P32:%.*]] = add i32 [[P30]], undef
-; DEFAULT-NEXT:    [[TMP3:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> [[TMP2]])
+; DEFAULT-NEXT:    [[TMP3:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> [[TMP2]])
 ; DEFAULT-NEXT:    [[OP_EXTRA]] = add i32 [[TMP3]], [[P17]]
 ; DEFAULT-NEXT:    [[P34:%.*]] = add i32 [[P32]], undef
 ; DEFAULT-NEXT:    br label [[FOR_BODY]]
@@ -76,7 +76,7 @@ define void @PR28330(i32 %n) {
 ; GATHER-NEXT:    [[TMP32:%.*]] = insertelement <8 x i32> [[TMP31]], i32 [[TMP25]], i32 6
 ; GATHER-NEXT:    [[TMP33:%.*]] = extractelement <8 x i32> [[TMP18]], i32 7
 ; GATHER-NEXT:    [[TMP34:%.*]] = insertelement <8 x i32> [[TMP32]], i32 [[TMP33]], i32 7
-; GATHER-NEXT:    [[TMP35:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> [[TMP34]])
+; GATHER-NEXT:    [[TMP35:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> [[TMP34]])
 ; GATHER-NEXT:    [[OP_EXTRA]] = add i32 [[TMP35]], [[P17]]
 ; GATHER-NEXT:    [[P34:%.*]] = add i32 [[P32]], [[TMP33]]
 ; GATHER-NEXT:    br label [[FOR_BODY]]
@@ -176,7 +176,7 @@ define void @PR32038(i32 %n) {
 ; DEFAULT-NEXT:    [[P28:%.*]] = add i32 [[P26]], undef
 ; DEFAULT-NEXT:    [[P30:%.*]] = add i32 [[P28]], undef
 ; DEFAULT-NEXT:    [[P32:%.*]] = add i32 [[P30]], undef
-; DEFAULT-NEXT:    [[TMP3:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> [[TMP2]])
+; DEFAULT-NEXT:    [[TMP3:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> [[TMP2]])
 ; DEFAULT-NEXT:    [[OP_EXTRA]] = add i32 [[TMP3]], -5
 ; DEFAULT-NEXT:    [[P34:%.*]] = add i32 [[P32]], undef
 ; DEFAULT-NEXT:    br label [[FOR_BODY]]
@@ -228,7 +228,7 @@ define void @PR32038(i32 %n) {
 ; GATHER-NEXT:    [[TMP32:%.*]] = insertelement <8 x i32> [[TMP31]], i32 [[TMP25]], i32 6
 ; GATHER-NEXT:    [[TMP33:%.*]] = extractelement <8 x i32> [[TMP18]], i32 7
 ; GATHER-NEXT:    [[TMP34:%.*]] = insertelement <8 x i32> [[TMP32]], i32 [[TMP33]], i32 7
-; GATHER-NEXT:    [[TMP35:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> [[TMP34]])
+; GATHER-NEXT:    [[TMP35:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> [[TMP34]])
 ; GATHER-NEXT:    [[OP_EXTRA]] = add i32 [[TMP35]], -5
 ; GATHER-NEXT:    [[P34:%.*]] = add i32 [[P32]], [[TMP33]]
 ; GATHER-NEXT:    br label [[FOR_BODY]]
@@ -266,7 +266,7 @@ define void @PR32038(i32 %n) {
 ; MAX-COST-NEXT:    [[P27:%.*]] = select i1 [[P9]], i32 -720, i32 -80
 ; MAX-COST-NEXT:    [[P28:%.*]] = add i32 [[P26]], [[P27]]
 ; MAX-COST-NEXT:    [[P29:%.*]] = select i1 [[P11]], i32 -720, i32 -80
-; MAX-COST-NEXT:    [[TMP9:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> [[TMP8]])
+; MAX-COST-NEXT:    [[TMP9:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP8]])
 ; MAX-COST-NEXT:    [[TMP10:%.*]] = add i32 [[TMP9]], [[P27]]
 ; MAX-COST-NEXT:    [[TMP11:%.*]] = add i32 [[TMP10]], [[P29]]
 ; MAX-COST-NEXT:    [[OP_EXTRA:%.*]] = add i32 [[TMP11]], -5

test/Transforms/SLPVectorizer/AArch64/horizontal.ll

@@ -49,7 +49,7 @@ define i32 @test_select(i32* noalias nocapture readonly %blk1, i32* noalias noca
 ; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 undef, [[S_026]]
 ; CHECK-NEXT:    [[ADD11:%.*]] = add nsw i32 [[ADD]], undef
 ; CHECK-NEXT:    [[ADD19:%.*]] = add nsw i32 [[ADD11]], undef
-; CHECK-NEXT:    [[TMP8:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> [[TMP7]])
+; CHECK-NEXT:    [[TMP8:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP7]])
 ; CHECK-NEXT:    [[OP_EXTRA]] = add nsw i32 [[TMP8]], [[S_026]]
 ; CHECK-NEXT:    [[ADD27:%.*]] = add nsw i32 [[ADD19]], undef
 ; CHECK-NEXT:    [[ADD_PTR]] = getelementptr inbounds i32, i32* [[P1_023]], i64 [[IDX_EXT]]
@@ -176,7 +176,7 @@ define i32 @reduction_with_br(i32* noalias nocapture readonly %blk1, i32* noalia
 ; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 undef, [[S_020]]
 ; CHECK-NEXT:    [[ADD5:%.*]] = add nsw i32 [[ADD]], undef
 ; CHECK-NEXT:    [[ADD9:%.*]] = add nsw i32 [[ADD5]], undef
-; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> [[TMP4]])
+; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP4]])
 ; CHECK-NEXT:    [[OP_EXTRA]] = add nsw i32 [[TMP5]], [[S_020]]
 ; CHECK-NEXT:    [[ADD13:%.*]] = add nsw i32 [[ADD9]], undef
 ; CHECK-NEXT:    [[CMP14:%.*]] = icmp slt i32 [[OP_EXTRA]], [[LIM:%.*]]
@@ -300,7 +300,7 @@ define i32 @test_unrolled_select(i8* noalias nocapture readonly %blk1, i8* noali
 ; CHECK-NEXT:    [[ADD49:%.*]] = add nsw i32 [[ADD38]], undef
 ; CHECK-NEXT:    [[ADD60:%.*]] = add nsw i32 [[ADD49]], undef
 ; CHECK-NEXT:    [[ADD71:%.*]] = add nsw i32 [[ADD60]], undef
-; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> [[TMP9]])
+; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> [[TMP9]])
 ; CHECK-NEXT:    [[OP_EXTRA]] = add nsw i32 [[TMP10]], [[S_047]]
 ; CHECK-NEXT:    [[ADD82:%.*]] = add nsw i32 [[ADD71]], undef
 ; CHECK-NEXT:    [[CMP83:%.*]] = icmp slt i32 [[OP_EXTRA]], [[LIM:%.*]]

test/Transforms/SLPVectorizer/AArch64/transpose.ll

@@ -260,7 +260,7 @@ define i32 @reduction_v4i32(<4 x i32> %v0, <4 x i32> %v1) {
 ; CHECK-NEXT:    [[TMP14:%.*]] = mul nuw <4 x i32> [[TMP13]], <i32 65535, i32 65535, i32 65535, i32 65535>
 ; CHECK-NEXT:    [[TMP15:%.*]] = add <4 x i32> [[TMP14]], [[TMP11]]
 ; CHECK-NEXT:    [[TMP16:%.*]] = xor <4 x i32> [[TMP15]], [[TMP14]]
-; CHECK-NEXT:    [[TMP17:%.*]] = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> [[TMP16]])
+; CHECK-NEXT:    [[TMP17:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP16]])
 ; CHECK-NEXT:    ret i32 [[TMP17]]
 ;
   %v0.0 = extractelement <4 x i32> %v0, i32 0

utils/TableGen/IntrinsicEmitter.cpp

@@ -219,7 +219,8 @@ enum IIT_Info {
   IIT_STRUCT6 = 38,
   IIT_STRUCT7 = 39,
   IIT_STRUCT8 = 40,
-  IIT_F128 = 41
+  IIT_F128 = 41,
+  IIT_VEC_ELEMENT = 42
 };
 
 static void EncodeFixedValueType(MVT::SimpleValueType VT,
@@ -289,6 +290,8 @@ static void EncodeFixedType(Record *R, std::vector<unsigned char> &ArgCodes,
       return;
     } else if (R->isSubClassOf("LLVMPointerToElt"))
       Sig.push_back(IIT_PTR_TO_ELT);
+    else if (R->isSubClassOf("LLVMVectorElementType"))
+      Sig.push_back(IIT_VEC_ELEMENT);
     else
       Sig.push_back(IIT_ARG);
     return Sig.push_back((Number << 3) | 7 /*IITDescriptor::AK_MatchType*/);