[SLP]Fix a crash when trying to reduce in revec after minbitwidth analysis

Need to use the original scalar type, when building the reduction, and
use the scalar type, when performing casting, to avoid compiler crash.

Author: alexey-bataev

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

@@ -22141,8 +22141,8 @@ class HorizontalReduction {
         if (isa<FixedVectorType>(ScalarTy)) {
           assert(SLPReVec && "FixedVectorType is not expected.");
           unsigned ScalarTyNumElements = getNumElements(ScalarTy);
-          Value *ReducedSubTree = PoisonValue::get(getWidenedType(
-              VectorizedRoot->getType()->getScalarType(), ScalarTyNumElements));
+          Value *ReducedSubTree = PoisonValue::get(
+              getWidenedType(ScalarTy->getScalarType(), ScalarTyNumElements));
           for (unsigned I : seq<unsigned>(ScalarTyNumElements)) {
             // Do reduction for each lane.
             // e.g., do reduce add for
@@ -22359,7 +22359,7 @@ class HorizontalReduction {
                         Type *DestTy) {
     Value *Rdx = emitReduction(Vec, Builder, &TTI, DestTy);
     if (Rdx->getType() != DestTy->getScalarType())
-      Rdx = Builder.CreateIntCast(Rdx, DestTy, IsSigned);
+      Rdx = Builder.CreateIntCast(Rdx, DestTy->getScalarType(), IsSigned);
     // Improved analysis for add/fadd/xor reductions with same scale
     // factor for all operands of reductions. We can emit scalar ops for
     // them instead.

llvm/test/Transforms/SLPVectorizer/SystemZ/revec-fix-128169.ll

@@ -1,5 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt -mtriple=s390x-unknown-linux-gnu -mcpu=z17 -passes=slp-vectorizer -S -slp-revec %s | FileCheck %s
+; RUN: opt -mtriple=s390x-unknown-linux-gnu -mcpu=z17 -passes=slp-vectorizer -S -slp-revec < %s | FileCheck %s
+; RUN: opt -mtriple=s390x-unknown-linux-gnu -mcpu=z17 -passes=slp-vectorizer -S -slp-revec -slp-threshold=-1000 < %s | FileCheck %s --check-prefix=THRESH
 
 define void @e(<4 x i16> %0) {
 ; CHECK-LABEL: @e(
@@ -22,6 +23,40 @@ define void @e(<4 x i16> %0) {
 ; CHECK-NEXT:    [[TMP12]] = or <4 x i32> [[TMP9]], [[TMP11]]
 ; CHECK-NEXT:    br label [[VECTOR_BODY]]
 ;
+; THRESH-LABEL: @e(
+; THRESH-NEXT:  entry:
+; THRESH-NEXT:    [[TMP1:%.*]] = call <8 x i16> @llvm.vector.insert.v8i16.v4i16(<8 x i16> poison, <4 x i16> zeroinitializer, i64 0)
+; THRESH-NEXT:    [[TMP2:%.*]] = call <8 x i16> @llvm.vector.insert.v8i16.v4i16(<8 x i16> [[TMP1]], <4 x i16> zeroinitializer, i64 4)
+; THRESH-NEXT:    [[TMP3:%.*]] = call <8 x i16> @llvm.vector.insert.v8i16.v4i16(<8 x i16> poison, <4 x i16> zeroinitializer, i64 4)
+; THRESH-NEXT:    [[TMP4:%.*]] = call <16 x i16> @llvm.vector.insert.v16i16.v4i16(<16 x i16> poison, <4 x i16> zeroinitializer, i64 0)
+; THRESH-NEXT:    [[TMP5:%.*]] = call <16 x i16> @llvm.vector.insert.v16i16.v4i16(<16 x i16> [[TMP4]], <4 x i16> zeroinitializer, i64 4)
+; THRESH-NEXT:    [[TMP6:%.*]] = call <16 x i16> @llvm.vector.insert.v16i16.v4i16(<16 x i16> [[TMP5]], <4 x i16> zeroinitializer, i64 8)
+; THRESH-NEXT:    [[TMP7:%.*]] = call <16 x i16> @llvm.vector.insert.v16i16.v4i16(<16 x i16> [[TMP6]], <4 x i16> zeroinitializer, i64 12)
+; THRESH-NEXT:    br label [[VECTOR_BODY:%.*]]
+; THRESH:       vector.body:
+; THRESH-NEXT:    [[VEC_IND:%.*]] = phi <4 x i16> [ zeroinitializer, [[ENTRY:%.*]] ], [ zeroinitializer, [[VECTOR_BODY]] ]
+; THRESH-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[ENTRY]] ], [ [[TMP26:%.*]], [[VECTOR_BODY]] ]
+; THRESH-NEXT:    [[TMP8:%.*]] = call <8 x i16> @llvm.vector.insert.v8i16.v4i16(<8 x i16> [[TMP3]], <4 x i16> [[VEC_IND]], i64 0)
+; THRESH-NEXT:    [[TMP9:%.*]] = add <8 x i16> [[TMP2]], [[TMP8]]
+; THRESH-NEXT:    [[TMP10:%.*]] = shufflevector <8 x i16> [[TMP8]], <8 x i16> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+; THRESH-NEXT:    [[TMP11:%.*]] = call <16 x i16> @llvm.vector.insert.v16i16.v4i16(<16 x i16> [[TMP10]], <4 x i16> [[TMP0:%.*]], i64 4)
+; THRESH-NEXT:    [[TMP12:%.*]] = call <16 x i16> @llvm.vector.insert.v16i16.v8i16(<16 x i16> [[TMP11]], <8 x i16> [[TMP9]], i64 8)
+; THRESH-NEXT:    [[TMP13:%.*]] = icmp sgt <16 x i16> [[TMP12]], [[TMP7]]
+; THRESH-NEXT:    [[TMP14:%.*]] = shufflevector <16 x i1> [[TMP13]], <16 x i1> poison, <4 x i32> <i32 0, i32 4, i32 8, i32 12>
+; THRESH-NEXT:    [[TMP15:%.*]] = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> [[TMP14]])
+; THRESH-NEXT:    [[TMP16:%.*]] = shufflevector <16 x i1> [[TMP13]], <16 x i1> poison, <4 x i32> <i32 1, i32 5, i32 9, i32 13>
+; THRESH-NEXT:    [[TMP17:%.*]] = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> [[TMP16]])
+; THRESH-NEXT:    [[TMP18:%.*]] = shufflevector <16 x i1> [[TMP13]], <16 x i1> poison, <4 x i32> <i32 2, i32 6, i32 10, i32 14>
+; THRESH-NEXT:    [[TMP19:%.*]] = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> [[TMP18]])
+; THRESH-NEXT:    [[TMP20:%.*]] = shufflevector <16 x i1> [[TMP13]], <16 x i1> poison, <4 x i32> <i32 3, i32 7, i32 11, i32 15>
+; THRESH-NEXT:    [[TMP21:%.*]] = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> [[TMP20]])
+; THRESH-NEXT:    [[TMP22:%.*]] = insertelement <4 x i1> poison, i1 [[TMP15]], i32 0
+; THRESH-NEXT:    [[TMP23:%.*]] = insertelement <4 x i1> [[TMP22]], i1 [[TMP17]], i32 1
+; THRESH-NEXT:    [[TMP24:%.*]] = insertelement <4 x i1> [[TMP23]], i1 [[TMP19]], i32 2
+; THRESH-NEXT:    [[TMP25:%.*]] = insertelement <4 x i1> [[TMP24]], i1 [[TMP21]], i32 3
+; THRESH-NEXT:    [[TMP26]] = zext <4 x i1> [[TMP25]] to <4 x i32>
+; THRESH-NEXT:    br label [[VECTOR_BODY]]
+;
 entry:
   br label %vector.body