[AArch64] Add a cost for v2i32 vecreduce.add.

These can lower to a addp. The score does not alter with this patch, but this
should help keep the scores the same with #146526.

Author: davemgreen

llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp

@@ -5205,6 +5205,7 @@ AArch64TTIImpl::getArithmeticReductionCost(unsigned Opcode, VectorType *ValTy,
       {ISD::ADD, MVT::v16i8,  2},
       {ISD::ADD, MVT::v4i16,  2},
       {ISD::ADD, MVT::v8i16,  2},
+      {ISD::ADD, MVT::v2i32,  2},
       {ISD::ADD, MVT::v4i32,  2},
       {ISD::ADD, MVT::v2i64,  2},
       {ISD::OR,  MVT::v8i8,  15},

llvm/test/Analysis/CostModel/AArch64/reduce-add.ll

@@ -6,6 +6,7 @@ target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
 define void @reduce() {
 ; CHECK-LABEL: 'reduce'
 ; CHECK-NEXT:  Cost Model: Found costs of 2 for: %V1i8 = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found costs of 2 for: %V2i8 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found costs of 2 for: %V3i8 = call i8 @llvm.vector.reduce.add.v3i8(<3 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found costs of 2 for: %V4i8 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found costs of 2 for: %V8i8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
@@ -24,6 +25,7 @@ define void @reduce() {
 ; CHECK-NEXT:  Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void
 ;
   %V1i8 = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> undef)
+  %V2i8 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
   %V3i8 = call i8 @llvm.vector.reduce.add.v3i8(<3 x i8> undef)
   %V4i8 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
   %V8i8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
@@ -41,20 +43,3 @@ define void @reduce() {
   %V4i64 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
   ret void
 }
-
-declare i8 @llvm.vector.reduce.add.v1i8(<1 x i8>)
-declare i8 @llvm.vector.reduce.add.v3i8(<3 x i8>)
-declare i8 @llvm.vector.reduce.add.v4i8(<4 x i8>)
-declare i8 @llvm.vector.reduce.add.v8i8(<8 x i8>)
-declare i8 @llvm.vector.reduce.add.v16i8(<16 x i8>)
-declare i8 @llvm.vector.reduce.add.v32i8(<32 x i8>)
-declare i8 @llvm.vector.reduce.add.v64i8(<64 x i8>)
-declare i16 @llvm.vector.reduce.add.v2i16(<2 x i16>)
-declare i16 @llvm.vector.reduce.add.v4i16(<4 x i16>)
-declare i16 @llvm.vector.reduce.add.v8i16(<8 x i16>)
-declare i16 @llvm.vector.reduce.add.v16i16(<16 x i16>)
-declare i32 @llvm.vector.reduce.add.v2i32(<2 x i32>)
-declare i32 @llvm.vector.reduce.add.v4i32(<4 x i32>)
-declare i32 @llvm.vector.reduce.add.v8i32(<8 x i32>)
-declare i64 @llvm.vector.reduce.add.v2i64(<2 x i64>)
-declare i64 @llvm.vector.reduce.add.v4i64(<4 x i64>)