[SLP]Fix PR77916: transform the whole mask, not only the elements for

the second vector.

Need to transform all elements in the long mask, if we decided to
produce shorter version, some elements may still have incorrect inifices
after transformation for the first vector in the permutation.

Author: alexey-bataev

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

@@ -7379,6 +7379,8 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
             continue;
           if (Idx >= static_cast<int>(CommonVF))
             Idx = E1Mask[Idx - CommonVF] + VF;
+          else
+            Idx = E1Mask[Idx];
         }
         CommonVF = VF;
       }

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

@@ -0,0 +1,45 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux -slp-threshold=-10 < %s | FileCheck %s
+
+define i32 @test() {
+; CHECK-LABEL: define i32 @test() {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    br label [[BB1:%.*]]
+; CHECK:       bb1:
+; CHECK-NEXT:    [[TMP0:%.*]] = phi <2 x i32> [ [[TMP5:%.*]], [[BB3:%.*]] ], [ zeroinitializer, [[BB:%.*]] ]
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <2 x i32> [[TMP0]], <2 x i32> poison, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1>
+; CHECK-NEXT:    br i1 false, label [[BB4:%.*]], label [[BB3]]
+; CHECK:       bb3:
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <2 x i32> [[TMP0]], <2 x i32> <i32 0, i32 poison>, <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[TMP3:%.*]] = add <2 x i32> zeroinitializer, [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = or <2 x i32> zeroinitializer, [[TMP2]]
+; CHECK-NEXT:    [[TMP5]] = shufflevector <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> <i32 0, i32 3>
+; CHECK-NEXT:    br label [[BB1]]
+; CHECK:       bb4:
+; CHECK-NEXT:    [[TMP6:%.*]] = phi <8 x i32> [ [[TMP1]], [[BB1]] ]
+; CHECK-NEXT:    ret i32 0
+;
+bb:
+  br label %bb1
+
+bb1:
+  %phi = phi i32 [ %or, %bb3 ], [ 0, %bb ]
+  %phi2 = phi i32 [ %add, %bb3 ], [ 0, %bb ]
+  br i1 false, label %bb4, label %bb3
+
+bb3:
+  %or = or i32 0, %phi
+  %add = add i32 0, 0
+  br label %bb1
+
+bb4:
+  %phi5 = phi i32 [ %phi2, %bb1 ]
+  %phi6 = phi i32 [ %phi2, %bb1 ]
+  %phi7 = phi i32 [ %phi2, %bb1 ]
+  %phi8 = phi i32 [ %phi2, %bb1 ]
+  %phi9 = phi i32 [ %phi2, %bb1 ]
+  %phi10 = phi i32 [ %phi2, %bb1 ]
+  %phi11 = phi i32 [ %phi, %bb1 ]
+  %phi12 = phi i32 [ %phi, %bb1 ]
+  ret i32 0
+}