diff --git a/csrc/codegen.cpp b/csrc/codegen.cpp
index aaddfe30071..97599504774 100644
--- a/csrc/codegen.cpp
+++ b/csrc/codegen.cpp
@@ -253,7 +253,7 @@ class CudaKernelGenerator : private kir::ConstIrVisitor {
   // array. This avoid the type mismatch in template functions when one of the
   // arguments is an aligned array (Array<T,N>) while the other is a regular
   // array T[N].
-  std::string genVarForTemplateFunction(Val* v) {
+  std::string genVariableNameConvertAlignedArray(Val* v) {
     TensorView* tv = nullptr;
     if (v->isA<kir::TensorIndex>()) {
       tv = v->as<kir::TensorIndex>()->view();
@@ -2318,13 +2318,14 @@ class CudaKernelGenerator : private kir::ConstIrVisitor {
     ArgumentBuilder func_args;
 
     // outputs
-    func_args.arg(genVarForTemplateFunction(output.get(0)));
-    func_args.arg(genVarForTemplateFunction(output.get(1)));
-    func_args.arg(genVarForTemplateFunction(output.get(2)));
+    func_args.arg(genVariableNameConvertAlignedArray(output.get(0)));
+    func_args.arg(genVariableNameConvertAlignedArray(output.get(1)));
+    func_args.arg(genVariableNameConvertAlignedArray(output.get(2)));
     // inputs
-    func_args.arg(genVarForTemplateFunction(input.get(0)));
-    func_args.arg(genVarForTemplateFunction(input.get(1)));
-    func_args.arg(genVarForTemplateFunction(input.get(2))).append("[0]");
+    func_args.arg(genVariableNameConvertAlignedArray(input.get(0)));
+    func_args.arg(genVariableNameConvertAlignedArray(input.get(1)));
+    func_args.arg(genVariableNameConvertAlignedArray(input.get(2)))
+        .append("[0]");
 
     // global buf
     for (const auto i : c10::irange(3)) {