[SLP]Use original vector if need to shuffle truncated root.

[ABataev]

Summary

If the root scalar is mapped to to the smallest bit width, the vector is
truncated and the types between original buildvector and extracted value
mismatched. For extract, we emit sext/zext instructions, for shuffles we
can reuse oringal vector instead of the truncated one.

[scui]

LGTM. Thanks for the quick fix!

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

for (const auto &ExternalUse : ExternalUses) {
     }
 
     if (auto *VU = dyn_cast<InsertElementInst>(User)) {
       // Skip if the scalar is another vector op or Vec is not an instruction.
       if (!Scalar->getType()->isVectorTy() && isa<Instruction>(Vec)) {
         if (auto *FTy = dyn_cast<FixedVectorType>(User->getType())) {
           Optional<unsigned> InsertIdx = getInsertIndex(VU);
           if (InsertIdx) {
+            // Need to use original vector, if the root is truncated.
+            if (MinBWs.count(Scalar) &&
+                VectorizableTree[0]->VectorizedValue == Vec)
+              Vec = VectorRoot;
             auto *It =
                 find_if(ShuffledInserts, [VU](const ShuffledInsertData &Data) {
                   // Checks if 2 insertelements are from the same buildvector.
                   InsertElementInst *VecInsert = Data.InsertElements.front();
                   return areTwoInsertFromSameBuildVector(VU, VecInsert);
                 });
             unsigned Idx = *InsertIdx;
             if (It == ShuffledInserts.end()) {

llvm/test/Transforms/SLPVectorizer/X86/root-trunc-extract-reuse.ll

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -slp-vectorizer -S -slp-threshold=-100 -mtriple=x86_64 < %s | FileCheck %s
+
+define i1 @test() {
+; CHECK-LABEL: @test(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK:       then:
+; CHECK-NEXT:    br label [[ELSE]]
+; CHECK:       else:
+; CHECK-NEXT:    [[TMP0:%.*]] = phi <2 x i32> [ zeroinitializer, [[THEN]] ], [ zeroinitializer, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc <2 x i32> [[TMP0]] to <2 x i8>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i8> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = zext i8 [[TMP2]] to i32
+; CHECK-NEXT:    [[BF_CAST162:%.*]] = and i32 [[TMP3]], 0
+; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <2 x i32> zeroinitializer, <2 x i32> [[TMP0]], <2 x i32> <i32 3, i32 1>
+; CHECK-NEXT:    [[T13:%.*]] = and <2 x i32> [[TMP4]], zeroinitializer
+; CHECK-NEXT:    br label [[ELSE1:%.*]]
+; CHECK:       else1:
+; CHECK-NEXT:    [[T20:%.*]] = extractelement <2 x i32> [[T13]], i64 0
+; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <2 x i32> poison, i32 [[BF_CAST162]], i32 0
+; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <2 x i32> [[TMP5]], i32 [[T20]], i32 1
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp ugt <2 x i32> [[TMP6]], zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i1> [[TMP7]], i32 1
+; CHECK-NEXT:    ret i1 [[TMP8]]
+;
+entry:
+  br i1 false, label %then, label %else
+
+then:
+  br label %else
+
+else:
+  %bf.load.off43 = phi i32 [ 0, %then ], [ 0, %entry ]
+  %bf.load.off44 = phi i32 [ 0, %then ], [ 0, %entry ]
+  %bf.cast162 = and i32 %bf.load.off43, 0
+  %t12 = insertelement <2 x i32> zeroinitializer, i32 %bf.load.off44, i64 0
+  %t13 = and <2 x i32> %t12, zeroinitializer
+  br label %else1
+
+else1:
+  %cmp40 = icmp ugt i32 %bf.cast162, 0
+  %t20 = extractelement <2 x i32> %t13, i64 0
+  %cmp50 = icmp ugt i32 %t20, 0
+  ret i1 %cmp50
+}