func.func private @simple_callee(%arg0: tensor<2xi32>) -> tensor<2xi32> {
  %graph = tf_executor.graph {
    tf_executor.fetch %arg0 : tensor<2xi32>
  }
  func.return %graph#0 : tensor<2xi32>
}

// Test with a TPUPartitionedCallOp that is not inlined.
func.func @dont_inline_tpu_partitioned_call(%arg0: tensor<2xi32> {tf._user_specified_name = "inputs_0"}, %arg1: tensor<2xi32> {tf._user_specified_name = "inputs_1"}) -> tensor<2xi32> {

//RUN: tf_tfl_translate --enable-stablehlo-conversion --input-mlir %s -o /tmp/temp.stablehlo; [ -f /tmp/temp.stablehlo ]


module {
func.func @main(%arg0: tensor<2xi32>) -> tensor<2xi32> {
  %0 = "tf.Div"(%arg0, %arg0) : (tensor<2xi32>, tensor<2xi32>) -> tensor<2xi32>
  func.return %0 : tensor<2xi32>
}

module {
func.func @main(%arg0: tensor<2xi32>, %arg1: tensor<2xi32>, %arg2: tensor<2xi32>) -> tensor<2xi32> {
  %0 = "tfl.custom"(%arg0, %arg1, %arg2) {custom_code = "stablehlo.clamp", custom_option = #tfl<const_bytes : "0x00000100002401">} : (tensor<2xi32>, tensor<2xi32>, tensor<2xi32>) -> tensor<2xi32>
  func.return %0 : tensor<2xi32>
}
}

// CHECK:       module {

func.func @QuantizeTransposeConvWeightOnly(%arg0: tensor<32x4x4x128xf32>, %arg1: tensor<4xi32>) -> tensor<1x32x42x128xf32> {
  %w = arith.constant dense<127.0> : tensor<1x32x42x128xf32>
  %b = arith.constant dense<0.0> : tensor<1x32x42x128xf32>

  func.return %1 : tensor<2xi32>
}
}

// CHECK:       module {
// CHECK-NEXT:  func @main(%arg0: tensor<2xi32>) -> tensor<2xi32> attributes {tf.entry_function = {inputs = "arg0", outputs = "tfl.custom1"}} {
// CHECK-NEXT:    %0 = stablehlo.add %arg0, %arg0 : tensor<2xi32>
// CHECK-NEXT:    %1 = stablehlo.subtract %0, %arg0 : tensor<2xi32>
// CHECK-NEXT:    return %1 : tensor<2xi32>

module {
func.func @main(%arg0: tensor<2xi32>) -> tensor<2xi32> attributes {tf.entry_function = {inputs = "arg0", outputs = "tfl.custom1"}} {
  %0 = "tfl.custom"(%arg0, %arg0) {custom_code = "stablehlo.maximum", custom_option = #tfl<const_bytes : "0x00000100002401">} : (tensor<2xi32>, tensor<2xi32>) -> tensor<2xi32>
  func.return %0 : tensor<2xi32>
}
}

// CHECK:       module {

module {
func.func @main(%arg0: tensor<2xi32>) -> tensor<2xi32> attributes {tf.entry_function = {inputs = "arg0", outputs = "tfl.custom1"}} {
  %0 = "tfl.custom"(%arg0, %arg0) {custom_code = "stablehlo.multiply", custom_option = #tfl<const_bytes : "0x00000100002401">} : (tensor<2xi32>, tensor<2xi32>) -> tensor<2xi32>
  func.return %0 : tensor<2xi32>
}
}

// CHECK:       module {

  // Convert input: NCHW -> NHWC
  %0 = "tf.Const"() {value = dense<[0, 2, 3, 1]> : tensor<4xi32>} : () -> tensor<4xi32>
  %1 = "tf.Transpose"(%arg0, %0) : (tensor<1x3x32x32xf32>, tensor<4xi32>) -> tensor<1x32x32x3xf32>

  // Compute in NHWC
  %2 = "tf.Conv2D"(%1, %arg1)
    {
      data_format = "NHWC",
      padding = "SAME",
      strides = [1, 1, 1, 1],

    } : (tensor<10xf32>, tensor<10xf32>, tensor<6xi32>) -> (tensor<10xf32>, tensor<6xi32>)
    // CHECK: return %[[RESULT_CAST]], %[[RESULT]]#1 : tensor<10xf32>, tensor<6xi32>
    func.return %0#0, %0#1 : tensor<10xf32>, tensor<6xi32>
  }

  // 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(

module {
func.func @main(%arg0: tensor<2xi32>) -> tensor<2xi32> attributes {tf.entry_function = {inputs = "arg0", outputs = "tfl.custom1"}} {
  %0 = "tfl.custom"(%arg0, %arg0) {custom_code = "stablehlo.add", custom_option = #tfl<const_bytes : "0x00000100002401">} : (tensor<2xi32>, tensor<2xi32>) -> tensor<2xi32>
  func.return %0 : tensor<2xi32>
}
}

// CHECK:       module {