diff --git a/llvm/lib/Target/AIE/AIECombinerHelper.cpp b/llvm/lib/Target/AIE/AIECombinerHelper.cpp
index 2b168f92d5df..4eb3a73aeac5 100644
--- a/llvm/lib/Target/AIE/AIECombinerHelper.cpp
+++ b/llvm/lib/Target/AIE/AIECombinerHelper.cpp
@@ -1791,13 +1791,13 @@ bool llvm::matchShuffleToVSel(
 
   const LLT DstTy = MRI.getType(DstReg);
   const LLT Src1Ty = MRI.getType(Src1Reg);
-  if (Src1Ty.getSizeInBits() != 512)
+  if (Src1Ty.getSizeInBits() != 512 || Src1Ty == LLT::scalar(64))
     return false;
 
   const unsigned NumDstElems = DstTy.getNumElements();
   const unsigned NumSrcElems = Src1Ty.getNumElements();
-  assert(NumDstElems == NumSrcElems &&
-         "Expected same number of elements in dst and src vector types");
+  if (NumDstElems != NumSrcElems)
+    return false;
 
   // Check that the shuffle mask can be converted into VSel mask:
   // 1. The shuffle mask doesn't contain indices that correspond to the same
@@ -1805,18 +1805,43 @@ bool llvm::matchShuffleToVSel(
   // the i-th element from Src2 is used.
   // 2. The mask indices modulo the number of elements are in strictly ascending
   // order.
-  int PrevIdx = Mask[0] % NumSrcElems;
   const size_t NumElems = Mask.size();
-  for (unsigned I = 1; I < NumElems; I++) {
-    int CurrIdx = Mask[I] % NumSrcElems;
+  unsigned I = 0;
+  int PrevIdx = -1;
+  while (Mask[I] == -1 && I < NumElems) {
+    ++I;
+    ++PrevIdx;
+  }
+
+  // The mask contains only -1
+  if (I == NumElems)
+    return false;
+
+  int CurrIdx = Mask[I] % NumSrcElems;
+  if (CurrIdx <= PrevIdx)
+    return false;
+
+  PrevIdx = CurrIdx;
+  ++I;
+  for (; I < NumElems; ++I) {
+    if (Mask[I] == -1) {
+      ++PrevIdx;
+      continue;
+    }
+
+    CurrIdx = Mask[I] % NumSrcElems;
     if (CurrIdx <= PrevIdx)
       return false;
+    PrevIdx = CurrIdx;
   }
 
   // Create the mask
   unsigned long long DstMask = 0;
-  for (unsigned I = 0; I < NumElems; I++) {
+  for (I = 0; I < NumElems; I++) {
     int Idx = Mask[I];
+    if (Idx == -1)
+      continue;
+
     if (Idx >= (int)NumSrcElems) {
       unsigned long long ElemMask = 1 << I;
       DstMask |= ElemMask;
diff --git a/llvm/test/CodeGen/AIE/GlobalISel/prelegalizercombiner-shuffle-vector.mir b/llvm/test/CodeGen/AIE/GlobalISel/prelegalizercombiner-shuffle-vector.mir
index 844e900bdbb2..7f2129e63ae0 100644
--- a/llvm/test/CodeGen/AIE/GlobalISel/prelegalizercombiner-shuffle-vector.mir
+++ b/llvm/test/CodeGen/AIE/GlobalISel/prelegalizercombiner-shuffle-vector.mir
@@ -120,31 +120,86 @@ body:             |
     PseudoRET implicit $lr, implicit %0
 ...
 ---
+name: shuffle_vector_all_undef
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+body:             |
+  bb.1.entry:
+    liveins: $x2
+
+    ; CHECK-LABEL: name: shuffle_vector_all_undef
+    ; CHECK: liveins: $x2
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[DEF]](<16 x s32>)
+    %1:_(<16 x s32>) = COPY $x2
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
+    %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1)
+    PseudoRET implicit $lr, implicit %0
+...
+---
 name: shuffle_vector_vsel_valid_mask_ordered_indices
 alignment:       16
 exposesReturnsTwice: false
 legalized:       false
 body:             |
   bb.1.entry:
-    liveins: $r0, $wl4, $x2
+    liveins: $x2
 
     ; CHECK-LABEL: name: shuffle_vector_vsel_valid_mask_ordered_indices
-    ; CHECK: liveins: $r0, $wl4, $x2
+    ; CHECK: liveins: $x2
     ; CHECK-NEXT: {{  $}}
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
-    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<8 x s32>) = COPY $wl4
-    ; CHECK-NEXT: [[AIE_UNPAD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_AIE_UNPAD_VECTOR [[COPY1]](<8 x s32>)
-    ; CHECK-NEXT: [[AIE_PAD_VECTOR_UNDEF:%[0-9]+]]:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF [[AIE_UNPAD_VECTOR]](<4 x s32>)
-    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[AIE_PAD_VECTOR_UNDEF]], [[COPY]], 65520
-    ; CHECK-NEXT: $x0 = COPY [[AIE_VSEL]](<16 x s32>)
-    ; CHECK-NEXT: PseudoRET implicit $lr, implicit $x0
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[DEF]], [[COPY]], 65520
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[AIE_VSEL]](<16 x s32>)
     %1:_(<16 x s32>) = COPY $x2
-    %4:_(<8 x s32>) = COPY $wl4
-    %3:_(<4 x s32>) = G_AIE_UNPAD_VECTOR %4(<8 x s32>)
-    %8:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF %3(<4 x s32>)
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
     %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(0, 1, 2, 3, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-    $x0 = COPY %0(<16 x s32>)
-    PseudoRET implicit $lr, implicit $x0
+    PseudoRET implicit $lr, implicit %0
+...
+---
+name: shuffle_vector_vsel_valid_mask_ordered_indices_with_undef
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+body:             |
+  bb.1.entry:
+    liveins: $x2
+
+    ; CHECK-LABEL: name: shuffle_vector_vsel_valid_mask_ordered_indices_with_undef
+    ; CHECK: liveins: $x2
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[DEF]], [[COPY]], 65520
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[AIE_VSEL]](<16 x s32>)
+    %1:_(<16 x s32>) = COPY $x2
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
+    %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(-1, -1, -1, -1, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
+    PseudoRET implicit $lr, implicit %0
+...
+---
+name: shuffle_vector_vsel_valid_mask_with_undef
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+body:             |
+  bb.1.entry:
+    liveins: $x2
+
+    ; CHECK-LABEL: name: shuffle_vector_vsel_valid_mask_with_undef
+    ; CHECK: liveins: $x2
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[DEF]], [[COPY]], 64992
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[AIE_VSEL]](<16 x s32>)
+    %1:_(<16 x s32>) = COPY $x2
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
+    %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(0, 1, 2, -1, -1, 21, 22, 23, 24, -1, 26, 27, 28, 29, 30, 31)
+    PseudoRET implicit $lr, implicit %0
 ...
 ---
 name: shuffle_vector_vsel_valid_mask_mixed_indices
@@ -153,25 +208,61 @@ exposesReturnsTwice: false
 legalized:       false
 body:             |
   bb.1.entry:
-    liveins: $r0, $wl4, $x2
+    liveins: $x2
 
     ; CHECK-LABEL: name: shuffle_vector_vsel_valid_mask_mixed_indices
-    ; CHECK: liveins: $r0, $wl4, $x2
+    ; CHECK: liveins: $x2
     ; CHECK-NEXT: {{  $}}
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
-    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<8 x s32>) = COPY $wl4
-    ; CHECK-NEXT: [[AIE_UNPAD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_AIE_UNPAD_VECTOR [[COPY1]](<8 x s32>)
-    ; CHECK-NEXT: [[AIE_PAD_VECTOR_UNDEF:%[0-9]+]]:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF [[AIE_UNPAD_VECTOR]](<4 x s32>)
-    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[AIE_PAD_VECTOR_UNDEF]], [[COPY]], 65521
-    ; CHECK-NEXT: $x0 = COPY [[AIE_VSEL]](<16 x s32>)
-    ; CHECK-NEXT: PseudoRET implicit $lr, implicit $x0
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[DEF]], [[COPY]], 65521
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[AIE_VSEL]](<16 x s32>)
     %1:_(<16 x s32>) = COPY $x2
-    %4:_(<8 x s32>) = COPY $wl4
-    %3:_(<4 x s32>) = G_AIE_UNPAD_VECTOR %4(<8 x s32>)
-    %8:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF %3(<4 x s32>)
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
     %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(16, 1, 2, 3, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-    $x0 = COPY %0(<16 x s32>)
-    PseudoRET implicit $lr, implicit $x0
+    PseudoRET implicit $lr, implicit %0
+...
+---
+name: shuffle_vector_vsel_valid_mask_mixed_indices_with_undef
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+body:             |
+  bb.1.entry:
+    liveins: $x2
+
+    ; CHECK-LABEL: name: shuffle_vector_vsel_valid_mask_mixed_indices_with_undef
+    ; CHECK: liveins: $x2
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[AIE_VSEL:%[0-9]+]]:_(<16 x s32>) = G_AIE_VSEL [[DEF]], [[COPY]], 64465
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[AIE_VSEL]](<16 x s32>)
+    %1:_(<16 x s32>) = COPY $x2
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
+    %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(16, 1, -1, 3, 20, -1, 22, 23, 24, 25, -1, 27, 28, 29, 30, 31)
+    PseudoRET implicit $lr, implicit %0
+...
+---
+name: shuffle_vector_vsel_invalid_mask
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+body:             |
+  bb.1.entry:
+    liveins: $x2
+
+    ; CHECK-LABEL: name: shuffle_vector_vsel_invalid_mask
+    ; CHECK: liveins: $x2
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<16 x s32>) = G_SHUFFLE_VECTOR [[DEF]](<16 x s32>), [[COPY]], shufflemask(0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[SHUF]](<16 x s32>)
+    %1:_(<16 x s32>) = COPY $x2
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
+    %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)
+    PseudoRET implicit $lr, implicit %0
 ...
 ---
 name: shuffle_vector_vsel_invalid_mask_with_repeated_index
@@ -180,25 +271,19 @@ exposesReturnsTwice: false
 legalized:       false
 body:             |
   bb.1.entry:
-    liveins: $r0, $wl4, $x2
+    liveins: $x2
 
     ; CHECK-LABEL: name: shuffle_vector_vsel_invalid_mask_with_repeated_index
-    ; CHECK: liveins: $r0, $wl4, $x2
+    ; CHECK: liveins: $x2
     ; CHECK-NEXT: {{  $}}
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
-    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<8 x s32>) = COPY $wl4
-    ; CHECK-NEXT: [[AIE_UNPAD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_AIE_UNPAD_VECTOR [[COPY1]](<8 x s32>)
-    ; CHECK-NEXT: [[AIE_PAD_VECTOR_UNDEF:%[0-9]+]]:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF [[AIE_UNPAD_VECTOR]](<4 x s32>)
-    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<16 x s32>) = G_SHUFFLE_VECTOR [[AIE_PAD_VECTOR_UNDEF]](<16 x s32>), [[COPY]], shufflemask(0, 1, 2, 3, 16, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-    ; CHECK-NEXT: $x0 = COPY [[SHUF]](<16 x s32>)
-    ; CHECK-NEXT: PseudoRET implicit $lr, implicit $x0
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<16 x s32>) = G_SHUFFLE_VECTOR [[DEF]](<16 x s32>), [[COPY]], shufflemask(0, 1, 2, 3, 16, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[SHUF]](<16 x s32>)
     %1:_(<16 x s32>) = COPY $x2
-    %4:_(<8 x s32>) = COPY $wl4
-    %3:_(<4 x s32>) = G_AIE_UNPAD_VECTOR %4(<8 x s32>)
-    %8:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF %3(<4 x s32>)
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
     %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(0, 1, 2, 3, 16, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-    $x0 = COPY %0(<16 x s32>)
-    PseudoRET implicit $lr, implicit $x0
+    PseudoRET implicit $lr, implicit %0
 ...
 ---
 name: shuffle_vector_vsel_invalid_mask_with_unordered_indices
@@ -207,23 +292,38 @@ exposesReturnsTwice: false
 legalized:       false
 body:             |
   bb.1.entry:
-    liveins: $r0, $wl4, $x2
+    liveins: $x2
 
     ; CHECK-LABEL: name: shuffle_vector_vsel_invalid_mask_with_unordered_indices
-    ; CHECK: liveins: $r0, $wl4, $x2
+    ; CHECK: liveins: $x2
     ; CHECK-NEXT: {{  $}}
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<16 x s32>) = COPY $x2
-    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<8 x s32>) = COPY $wl4
-    ; CHECK-NEXT: [[AIE_UNPAD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_AIE_UNPAD_VECTOR [[COPY1]](<8 x s32>)
-    ; CHECK-NEXT: [[AIE_PAD_VECTOR_UNDEF:%[0-9]+]]:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF [[AIE_UNPAD_VECTOR]](<4 x s32>)
-    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<16 x s32>) = G_SHUFFLE_VECTOR [[AIE_PAD_VECTOR_UNDEF]](<16 x s32>), [[COPY]], shufflemask(1, 0, 2, 3, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-    ; CHECK-NEXT: $x0 = COPY [[SHUF]](<16 x s32>)
-    ; CHECK-NEXT: PseudoRET implicit $lr, implicit $x0
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<16 x s32>) = G_SHUFFLE_VECTOR [[DEF]](<16 x s32>), [[COPY]], shufflemask(1, 0, 2, 3, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[SHUF]](<16 x s32>)
     %1:_(<16 x s32>) = COPY $x2
-    %4:_(<8 x s32>) = COPY $wl4
-    %3:_(<4 x s32>) = G_AIE_UNPAD_VECTOR %4(<8 x s32>)
-    %8:_(<16 x s32>) = G_AIE_PAD_VECTOR_UNDEF %3(<4 x s32>)
+    %8:_(<16 x s32>) = G_IMPLICIT_DEF
     %0:_(<16 x s32>) = G_SHUFFLE_VECTOR %8(<16 x s32>), %1, shufflemask(1, 0, 2, 3, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
-    $x0 = COPY %0(<16 x s32>)
-    PseudoRET implicit $lr, implicit $x0
+    PseudoRET implicit $lr, implicit %0
+...
+---
+name: shuffle_vector_vsel_invalid_s64
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+body:             |
+  bb.1.entry:
+    liveins: $r0, $wl4, $x2
+
+    ; CHECK-LABEL: name: shuffle_vector_vsel_invalid_s64
+    ; CHECK: liveins: $r0, $wl4, $x2
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<8 x s64>) = COPY $x2
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<8 x s64>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<8 x s64>) = G_SHUFFLE_VECTOR [[DEF]](<8 x s64>), [[COPY]], shufflemask(8, 1, 2, 3, 11, 12, 13, 14)
+    ; CHECK-NEXT: PseudoRET implicit $lr, implicit [[SHUF]](<8 x s64>)
+    %1:_(<8 x s64>) = COPY $x2
+    %8:_(<8 x s64>) = G_IMPLICIT_DEF
+    %0:_(<8 x s64>) = G_SHUFFLE_VECTOR %8(<8 x s64>), %1, shufflemask(8, 1, 2, 3, 11, 12, 13, 14)
+    PseudoRET implicit $lr, implicit %0
 ...