[SLP]Use the size of gathered scalars when evaluating slice size

Need to use the size of the gathered scalars, not the original size of
the buildvector scalars, since gathered scalar size might be changed
during building the buildvector shuffles.

Fixes #125259

Author: alexey-bataev

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

@@ -15094,7 +15094,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Type *ScalarTy,
       }
     }
     if (!GatherShuffles.empty()) {
-      unsigned SliceSize = getPartNumElems(E->Scalars.size(), NumParts);
+      unsigned SliceSize = getPartNumElems(GatheredScalars.size(), NumParts);
       SmallVector<int> VecMask(Mask.size(), PoisonMaskElem);
       for (const auto [I, TEs] : enumerate(Entries)) {
         if (TEs.empty()) {

llvm/test/Transforms/SLPVectorizer/X86/gathered-shuffle-resized.ll

@@ -18,9 +18,9 @@ define ptr @test(ptr %0, ptr %args_gep) {
 ; CHECK-NEXT:    [[TMP8:%.*]] = load i32, ptr [[TMP7]], align 4, !noalias [[META0:![0-9]+]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = load <2 x i32>, ptr [[TMP6]], align 8, !noalias [[META0]]
 ; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
-; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i32> [[TMP9]], <2 x i32> poison, <16 x i32> poison
+; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i32> [[TMP9]], <2 x i32> poison, <16 x i32> <i32 poison, i32 poison, i32 0, i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
 ; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <4 x i32> [[TMP10]], <4 x i32> 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>
-; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP9]], <16 x i32> <i32 0, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP9]], <16 x i32> <i32 0, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 3, i32 3, i32 3>
 ; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq <16 x i32> [[TMP13]], zeroinitializer
 ; CHECK-NEXT:    [[TMP15:%.*]] = zext <16 x i1> [[TMP14]] to <16 x i8>
 ; CHECK-NEXT:    store <16 x i8> [[TMP15]], ptr [[TMP5]], align 1