diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -738,7 +738,7 @@
       return getShuffleCost(TTI::SK_ExtractSubvector, SrcTy, SubIndex, Ty);
 
     if (Shuffle->changesLength())
-      return -1;
+      BaseT::getUserCost(U, Operands);
 
     if (Shuffle->isIdentity())
       return 0;
diff --git a/llvm/test/Analysis/CostModel/AMDGPU/shufflevector.ll b/llvm/test/Analysis/CostModel/AMDGPU/shufflevector.ll
--- a/llvm/test/Analysis/CostModel/AMDGPU/shufflevector.ll
+++ b/llvm/test/Analysis/CostModel/AMDGPU/shufflevector.ll
@@ -51,3 +51,11 @@
   store <2 x i16> %shuf, <2 x i16> addrspace(1)* %out
   ret void
 }
+
+; assertion test
+define amdgpu_kernel void @shufflevector_xxx(<4 x i8> addrspace(1)* %out, <2 x i8> addrspace(1)* %vaddr) {
+  %vec = load <2 x i8>, <2 x i8> addrspace(1)* %vaddr
+  %shuf = shufflevector <2 x i8> %vec, <2 x i8> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  store <4 x i8> %shuf, <4 x i8> addrspace(1)* %out
+  ret void
+}