func.func @main(%arg0: tensor<*x!tf_type.resource>, %arg1: tensor<*x!tf_type.resource<tensor<3x3x1x32xf32>>>, %arg2: tensor<*xf32>, %arg3: tensor<2x4x6x8xi32>) -> (tensor<f32>, tensor<f32>)
attributes {tf.entry_function = {inputs = "args_0,args_1,args_2,args_3", outputs = "rets_0_RetVal,rets_1_RetVal"}} {
  %graph:2 = tf_executor.graph {

# CHECK-SAME: allow_soft_placement
# CHECK-SAME: control_outputs = ""
# CHECK-SAME: inputs = "args_0,args_1,args_2,args_3"
# CHECK-SAME: outputs = "rets_0,rets_1"
# CHECK:          %[[ISLAND_0:.*]], %[[ISLAND_0_control:.*]] = tf_executor.island wraps "tf.Const"
# CHECK:          %[[ISLAND_1:.*]], %[[ISLAND_1_control:.*]] = tf_executor.island wraps "tf.Identity"(%[[ISLAND_0]])

    // CHECK-NEXT: "tf.MakeIterator"(%[[ARG_0]], %[[ANON_ITER]])
    // CHECK-NEXT: %[[COND:.*]] = "tf.Const"
    // CHECK-NEXT: %[[WHILE:[0-9]*]]:5 = "tf.WhileRegion"(%[[COND]], %[[ARG_1]], %[[ARG_2]], %[[ARG_3]], %[[ARG_4]])
    // CHECK-NEXT:   ^bb0(%[[ARG_5:.*]]: tensor<i1>, %[[ARG_6:.*]]: tensor<i64>, %[[ARG_7:.*]]: tensor<i32>, %[[ARG_8:.*]]: tensor<!tf_type.resource<tensor<64xf32>>>, %[[ARG_9:.*]]: tensor<!tf_type.resource<tensor<128xf32>>>)

  func.func @main(%arg0: tensor<*xf32> {tf.device = "/CPU:0"}, %arg1: tensor<2x4x6x8xi32>) -> (tensor<*xf32>, tensor<2x4x6x8xi32> {tf.device = "/CPU:1"})
  attributes  {tf.entry_function = {inputs = "args_0,args_1", outputs = "rets_0,rets_1"}} {
    %0:2 = tf_executor.graph {

// Verify tf.aliasing_output attributes on args are dropped during export.

module attributes {tf.versions = {bad_consumers = [], min_consumer = 0 : i32, producer = 121 : i32}} {
  func.func @main(%arg0: tensor<*xf32> {tf.aliasing_output = 0 : i64}, %arg1: tensor<2x4x6x8xi32>) -> (tensor<*xf32>, tensor<2x4x6x8xi32> {tf.device = "/CPU:1"})
  attributes  {tf.entry_function = {inputs = "args_0,args_1", outputs = "rets_0,rets_1"}} {

        then:
        output1.toString().contains(ARG_1)
        output1.toString().contains(ARG_2)

        when:
        String output2 = runBuild { launcher -> launcher.addArguments(["-P$ARG_1" as String]).addArguments(["-P$ARG_2" as String]) }

        then:
        output2.toString().contains(ARG_1)
        output2.toString().contains(ARG_2)
    }

    value {
      type: DT_FLOAT
    }
  }
}
node {
  name: "args_0"
  op: "_Arg"
  attr {
    key: "T"
    value {
      type: DT_RESOURCE
    }
  }
  attr {
    key: "index"
    value {
      i: 0
    }
  }
}
node {
  name: "spc1"
  op: "StatefulPartitionedCall"
  input: "input0"
  input: "args_0"
  attr {
    key: "Tin"
    value {
      list {

  }

  // CHECK-LABEL: func private @main_0
  // CHECK-SAME: %[[ARG_0:.*]]: tensor<10xbf16>, %[[ARG_1:.*]]: tensor<10xbf16>, %[[ARG_2:.*]]: tensor<6xi32>
  func.func private @main_0(
      %arg0: tensor<10xf32>, %arg1: tensor<10xf32>, %arg2: tensor<6xi32>
    ) -> (tensor<10xf32>, tensor<6xi32>) attributes {_from_xla_call_module} {
    // CHECK: %[[ADD:.*]] = stablehlo.add %[[ARG_0]], %[[ARG_1]] : tensor<10xbf16>

        assertThat(args2.length, is(1));

        final ParameterizedClassDesc arg2_1 = args2[0];
        assertThat(arg2_1.getRawClass(), is(sameClass(Integer.class)));
    }

    /**
     * @throws Exception
     */
    @Test
    public void testMethodParameterType() throws Exception {
        final Map<TypeVariable<?>, Type> map = ParameterizedClassDescFactory.getTypeVariables(Hoge.class);

    func.return %4 : tensor<!tf_type.variant>
  }
  // CHECK-LABEL: __inference_Dataset_flat_map_lambda_190
  func.func private @__inference_Dataset_flat_map_lambda_190(%arg0: tensor<i64> {tf._user_specified_name = "args_0"}) -> tensor<!tf_type.variant> attributes {tf._original_func_name = "__inference_Dataset_flat_map_lambda_19", tf._tf_data_function = true, tf.signature.is_stateful} {