// PerTensor: "tfl.transpose_conv"(%arg1, %[[DEQUANTIZE]], %arg0,
}

// CHECK-LABEL: bias_adjust_pertensor
func.func @bias_adjust_pertensor(%arg0: tensor<1x2xf32>) -> (tensor<1x2xf32>) {
  %0 = "quantfork.stats"(%arg0) {
    layerStats = dense<[-1.28e-5, 1.27e-5]> : tensor<2xf32>
  } : (tensor<1x2xf32>) -> tensor<1x2xf32>

func.func @while_cond(%arg0: tensor<i32>) -> tensor<i1> {
  %0 = "tf.Const"() {value = dense<9> : tensor<i32>} : () -> tensor<i32>
  %1 = "tf.Less"(%arg0, %0) {} : (tensor<i32>, tensor<i32>) -> tensor<i1>
  func.return %1 : tensor<i1>
}

func.func @while_body(%arg0: tensor<i32>) -> tensor<i32> {
  %1 = "tf.AddV2"(%arg0, %arg0) {} : (tensor<i32>, tensor<i32>) -> tensor<i32>
  func.return %1 : tensor<i32>
}

func.func @broadcast_mul0(%arg0: tensor<5x7xf32>, %arg1: tensor<7xf32>) -> tensor<5x7xf32> {
  %cst = arith.constant dense<[5, 7]> : tensor<2xi32>
  %0 = "tf.BroadcastTo"(%arg1, %cst) : (tensor<7xf32>, tensor<2xi32>) -> tensor<5x7xf32>
  %1 = "tf.Mul"(%arg0, %0) : (tensor<5x7xf32>, tensor<5x7xf32>) -> tensor<5x7xf32>
  func.return %1 : tensor<5x7xf32>
  // CHECK: %[[V0:.*]] = "tf.Mul"(%arg0, %arg1) : (tensor<5x7xf32>, tensor<7xf32>) -> tensor<5x7xf32>

// A resource is passed into tf.If
func.func @cond_false(%arg0: tensor<!tf_type.resource<tensor<f32>>>, %arg1: tensor<f32>) -> tensor<f32> {
  func.return %arg1 : tensor<f32>
}

func.func @cond_true(%arg0: tensor<!tf_type.resource<tensor<f32>>>, %arg1: tensor<f32>) -> tensor<f32> {
  %0 = "tf.Const"() {value = dense<1.000000e+00> : tensor<f32>} : () -> tensor<f32>
  %1 = "tf.ReadVariableOp"(%arg0) : (tensor<!tf_type.resource<tensor<f32>>>) -> tensor<f32>

  func.return %0#0 : tensor<*xi32>
}

func.func private @compute_0(%arg0: tensor<?x?xi32> {tf._user_specified_name = "0"}, %arg1: tensor<?x?xi32>) -> (tensor<?x?xi32>, tensor<?x?xi32>) {
  func.return %arg0, %arg1 : tensor<?x?xi32>, tensor<?x?xi32>
}

func.func private @batch_3(%arg0: tensor<?x1xi32>) -> tensor<*xi32> {

//   id0 = arg0
//   if-then-branch: [u0,   arg0, arg0]
//   if-else-branch: [arg0, arg0, arg1]
//     => first result is unknown, second and third is passthrough
//   if results    : [*,    arg0, {arg0, arg1}[
//   ID #2: read (unknown)         -> succ {5, 6)
//   ID #3: read (arg0)            -> succ {5}
//   ID #4: read({arg0,arg1})      -> succ {5,6}
//   ID #5: write(arg0)

    %0 = "tf.MatMul"(%arg0, %arg1) {attr_map = "0:transpose_a,1:transpose_b", device = "", transpose_a = false, transpose_b = false} : (tensor<2x2xf32>, tensor<2x2xf32>) -> tensor<2x2xf32>
    return %0 : tensor<2x2xf32>
  }
  func.func private @composite_matmul_fn_2_0(%arg0: tensor<2x2xf32>, %arg1: tensor<2x2xf32>) -> tensor<2x2xf32> attributes {tf_quant.composite_function} {

// CHECK-NEXT: return %[[BACK]] : tensor<1x2x3xf32>
}

// CHECK-LABEL: pack_multiple
func.func @pack_multiple(%arg0: tensor<2x3xf32>,
                    %arg1: tensor<2x3xf32>,
                    %arg2: tensor<2x3xf32>) -> tensor<3x2x3xf32> {
  %0 = "tf.MyPack"(%arg0, %arg1, %arg2) {N=3:i32, axis=0:i32} : (tensor<2x3xf32>, tensor<2x3xf32>, tensor<2x3xf32>) -> tensor<3x2x3xf32>
  func.return %0 : tensor<3x2x3xf32>

    For example,

    ```mlir
      %0 = "tf.ReadVariableOp"(%arg0)
      %1 = "tf.ReadVariableOp"(%arg1)
      %2 = "tf.TPUExecute"(%0, %1, %compile)
      %3 = "tf.AssignVariableOp"(%arg0, %2)
    ```

    will be transformed into

    ```mlir
      %2 = "tf.TPUExecuteAndUpdateVariables"(%arg0, %arg1, %compile)
        { device_var_reads_indices = [0, 1],

  func.return %1 : tensor<1xf32>
}

func.func @cond_true(%arg0: tensor<*xf32>, %arg1: tensor<*xf32>) -> tensor<*xf32> {
  %0 = tfl.add %arg0, %arg1 {fused_activation_function = "NONE"} : tensor<*xf32>
  func.return %0 : tensor<*xf32>
}

func.func @cond_false(%arg0: tensor<*xf32>, %arg1: tensor<*xf32>) -> tensor<*xf32> {
  %0 = tfl.mul %arg0, %arg1 {fused_activation_function = "NONE"} : tensor<*xf32>
  func.return %0 : tensor<*xf32>