func.func @simple_chain_with_broadcast(%arg0: tensor<1xf32>, %arg1: tensor<10xf32>) -> tensor<?xf32> {
    // CHECK: %[[MUL:.*]] = "tf.Mul"{{.*}} (tensor<1xf32>, tensor<10xf32>) -> tensor<10xf32>
    // CHECK: %[[ADD:.*]] = "tf.Add"(%[[MUL]], %[[MUL]]) : (tensor<10xf32>, tensor<10xf32>) -> tensor<10xf32>
    // CHECK: %[[CAST:.*]] = "tf.Cast"(%[[ADD]]) {{.*}} : (tensor<10xf32>) -> tensor<*xf32>

module {
  // CHECK-LABEL: func @main
  // CHECK-SAME: %[[ARG_0:.*]]: tensor<10xf32>, %[[ARG_1:.*]]: tensor<10xf32>, %[[ARG_2:.*]]: tensor<6xi32>
  func.func @main(
      %arg0: tensor<10xf32>, %arg1: tensor<10xf32>, %arg2: tensor<6xi32>
    ) -> (tensor<10xf32>, tensor<6xi32>) {
    // CHECK: %[[CAST_0:.*]] = "tf.Cast"(%[[ARG_0]]) <{Truncate = false}> : (tensor<10xf32>) -> tensor<10xbf16>

    %0 = stablehlo.constant dense<500> : tensor<100xi64>
    %1 = stablehlo.constant dense<0> : tensor<100xi64>
    %2 = stablehlo.compare  LT, %arg2, %1 : (tensor<100xi64>, tensor<100xi64>) -> tensor<100xi1>
    %3 = stablehlo.add %arg2, %0 : tensor<100xi64>
    %4 = stablehlo.select %2, %3, %arg2 : tensor<100xi1>, tensor<100xi64>
    %5 = stablehlo.reshape %4 : (tensor<100xi64>) -> tensor<100x1xi64>

module{
  func.func @main(%arg0: tensor<10xf32>) -> tensor<10xf32> {
    return %arg0 : tensor<10xf32>
  }
  func.func @func1(%arg0: tensor<10xf32>, %arg1: tensor<10xf32>) -> tensor<10xf32> {
    %0 = stablehlo.add %arg0, %arg1 : tensor<10xf32>
    %1 = stablehlo.add %arg0, %arg1 : tensor<10xf32>
    return %0 : tensor<10xf32>
  }
})mlir";

// CHECK-DAG:       [[VAL_33:%.*]] = arith.constant dense<[-1, 0, 0]> : tensor<3xi32>...

  // CHECK: "tf.Conv3DBackpropInputV2"
}

func.func @mul_i64(%arg0: tensor<14xi64>, %arg1: tensor<14xi64>) -> tensor<14xi64> {
  %0 = "tf.Mul"(%arg0, %arg1) : (tensor<14xi64>, tensor<14xi64>) -> tensor<14xi64>
  func.return %0: tensor<14xi64>

// CHECK-LABEL: mul_i64
// CHECK:  tfl.mul %arg0, %arg1 {fused_activation_function = "NONE"} : tensor<14xi64>
// CHECK:  return
}

  %cst = "tf.Const"() {value = dense<0.000000e+00> : tensor<10xf32>} : () -> tensor<10xf32>
  %0 = "tf.MatMul"(%arg0, %arg1) {
    transpose_a = false, transpose_b = false
  } : (tensor<1x10xf32>, tensor<10x10xf32>) -> tensor<*xf32>
  %1 = "tf.BiasAdd"(%0, %cst) {data_format = "NHWC", device = ""} : (tensor<*xf32>, tensor<10xf32>) -> tensor<*xf32>
  %2 = "tf.Relu6"(%1) {device = ""} : (tensor<*xf32>) -> tensor<*xf32>

  %0 = mhlo.constant dense<0.000000e+00> : tensor<32x10xf32>
  %1 = mhlo.constant dense<0.000000e+00> : tensor<32x28x28x5xf32>
  %2 = "tf.ReadVariableOp"(%arg4) : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32>
  %3 = "tf.ReadVariableOp"(%arg2) : (tensor<!tf_type.resource<tensor<5xf32>>>) -> tensor<5xf32>

  %1 = "arith.constant"() {value = dense<0> : tensor<16xi64>} : () -> tensor<16xi64>
  %2 = "tfl.conv_2d"(%arg0, %0, %1) {dilation_h_factor = 1 : i32, dilation_w_factor = 1 : i32, fused_activation_function = "NONE", padding = "SAME", stride_h = 1 : i32, stride_w = 1 : i32} : (tensor<1x32x32x8x!quant.uniform<i16:f32, 1.0>>, tensor<16x1x1x8x!quant.uniform<i8:f32, 1.0>>, tensor<16xi64>) -> tensor<1x32x32x16x!quant.uniform<i16:f32, 1.0>>

func.func @testCompatibleCastType(%arg0: tensor<?xf32>) -> (tensor<10xf32>, tensor<10xf32>) {
  %0 = "tf.Cast"(%arg0) {Truncate = false} : (tensor<?xf32>) -> tensor<10xf32>
  %1 = "tf.Cast"(%arg0) {Truncate = true} : (tensor<?xf32>) -> tensor<10xf32>
  func.return %0, %1: tensor<10xf32>, tensor<10xf32>

  // CHECK: %0 = "tf.Cast"(%arg0) <{Truncate = false}> : (tensor<?xf32>) -> tensor<10xf32>