%0 = "tf.Const"() {value = dense<42> : tensor<1xi64>} : () -> tensor<1xi64>
  return
}
func.func @init2() attributes {tf_saved_model.exported_names = ["init2"]} {
  %0 = "tf.Const"() {value = dense<43> : tensor<1xi64>} : () -> tensor<1xi64>
  return
}
func.func @init3() attributes {tf_saved_model.exported_names = ["init3"]} {
  %0 = "tf.Const"() {value = dense<44> : tensor<1xi64>} : () -> tensor<1xi64>
  return
}

    %16 = "tf.BroadcastArgs"(%11, %14) : (tensor<1xi64>, tensor<1xi64>) -> tensor<1xi64>
    %17 = "tf.Concat"(%cst_2, %16, %12) : (tensor<i32>, tensor<1xi64>, tensor<2xi64>) -> tensor<3xi64>
    %18 = "tf.Concat"(%cst_2, %16, %15) : (tensor<i32>, tensor<1xi64>, tensor<2xi64>) -> tensor<3xi64>
    %19 = "tf.BroadcastTo"(%4, %17) : (tensor<2x?x?xi8>, tensor<3xi64>) -> tensor<2x?x?xi8>

module {
  func.func @main() -> tensor<1xi64> {
  %0 = "tfl.custom"() {custom_code = "stablehlo.constant", custom_option = #tfl<const_bytes : "0x76616C75650001020109010101062C022401">} : () -> tensor<1xi64>
  func.return %0 : tensor<1xi64>
  }
}

// CHECK:  module {
// CHECK-NEXT:    func @main() -> tensor<1xi64> {
// CHECK-NEXT:    %[[c0:.+]] = stablehlo.constant dense<2> : tensor<1xi64>
// CHECK-NEXT:    return %[[c0]] : tensor<1xi64>

  %cst = "tf.Const"() <{value = dense<0.000000e+00> : tensor<f32>}> : () -> tensor<f32>
  %cst_0 = "tf.Const"() <{value = dense<1> : tensor<1xi64>}> : () -> tensor<1xi64>
  %fill = "tf.Fill"(%cst_0, %cst) : (tensor<1xi64>, tensor<f32>) -> tensor<1xf32>
  tf_device.replicate([%0, %fill] as %arg_r0: tensor<1xf32>) {n = 2 : i32} {
    %1 = "tf_device.launch"() <{device = "TPU_REPLICATED_HOST_0"}> ({

  %0 = "vhlo.dot_general_v1"(%arg0, %arg1) <{
    lhs_batching_dimensions = #vhlo.tensor_v1<dense<0> : tensor<1xi64>>,
    lhs_contracting_dimensions = #vhlo.tensor_v1<dense<2> : tensor<1xi64>>,
    rhs_batching_dimensions = #vhlo.tensor_v1<dense<0> : tensor<1xi64>>,
    rhs_contracting_dimensions = #vhlo.tensor_v1<dense<0> : tensor<1xi64>>,

// CHECK-DAG:       %[[VAL_2:.*]] = "tf.Const"() <{value = dense<[1, 0]> : tensor<2xi64>}> : () -> tensor<2xi64>
// CHECK-DAG:       %[[VAL_3:.*]] = "tf.Const"() <{value = dense<[1, 0]> : tensor<2xi64>}> : () -> tensor<2xi64>
// CHECK:           %[[VAL_4:.*]] = "tf.Transpose"(%[[VAL_0]], %[[VAL_3]]) : (tensor<2x3xf32>, tensor<2xi64>) -> tensor<3x2xf32>

}

func.func @broadcast_args_i64(%arg0: tensor<3xi64>, %arg1: tensor<1xi64>) -> tensor<3xi64> {
  %0 = "tf.BroadcastArgs"(%arg0, %arg1) : (tensor<3xi64>, tensor<1xi64>) -> tensor<3xi64>
  func.return %0 : tensor<3xi64>

// CHECK-LABEL:broadcast_args_i64
// CHECK:  "tfl.broadcast_args"(%arg0, %arg1) : (tensor<3xi64>, tensor<1xi64>) -> tensor<3xi64>
}

  // NOT-CHECK-DAG:    %[[LHS_BCAST:.+]] = "mhlo.dynamic_broadcast_in_dim"(%arg0, %[[RESULT_EXTENTS]]) <{broadcast_dimensions = dense<0> : tensor<1xi64>}>
  // NOT-CHECK-DAG:    %[[RHS_BCAST:.+]] = "mhlo.dynamic_broadcast_in_dim"(%arg1, %[[RESULT_EXTENTS]]) <{broadcast_dimensions = dense<0> : tensor<1xi64>}>
  // NOT-CHECK-NEXT:   %[[RESULT:.+]] = mhlo.compare EQ, %[[LHS_BCAST]], %[[RHS_BCAST]]
  // NOT-CHECK-NEXT:   shape.assuming_yield %[[RESULT]]

    %dimension = "tf.Const"() { value = dense<3> : tensor<1xi64> } : () -> tensor<1xi64>
    %6 = "tf.Sum"(%3, %dimension) { keep_dims = true }: (tensor<1x3x2x2xf32>, tensor<1xi64>) -> tensor<1x3x2x1xf32>
    return %5, %6 : tensor<1x3x2x2xf32>, tensor<1x3x2x1xf32>
  }

// CHECK-LABEL: func @conv_with_multiple_uses

  // CHECK-DAG:  %[[CST0:.*]] = "tf.Const"() <{value = dense<[3, 2, 4]> : tensor<3xi64>}> : () -> tensor<3xi64>
  // CHECK-DAG:  %[[CST1:.*]] = "tf.Const"() <{value = dense<0> : tensor<3xi64>}> : () -> tensor<3xi64>
  // CHECK-DAG:  %[[CST2:.*]] = "tf.Const"() <{value = dense<[3, 6, 4]> : tensor<3xi64>}> : () -> tensor<3xi64>