%2 = "tfl.batch_matmul"(%1, %arg2) {adj_x = true, adj_y = false, asymmetric_quantize_inputs = false} : (tensor<4x8xf32>, tensor<4x256xf32>) -> tensor<8x256xf32>
  func.return %2 : tensor<8x256xf32>
  // CHECK: return %[[RES1]] : tensor<8x256xf32>
}

// CHECK-LABEL: Batchmatmul2Fullyconnected
// CHECK-NOT: "tfl.batch_matmul"

}

// CHECK-LABEL: QuantizeConcat
func.func @QuantizeConcat(tensor<1x2xf32>, tensor<1x2xf32>) -> tensor<2x2x!quant.uniform<u8:f32, 1.000000e-01:128>> {
^bb0(%arg0: tensor<1x2xf32>, %arg1: tensor<1x2xf32>):
  %0 = "tfl.concatenation"(%arg0, %arg1) {axis = 0 : i32, fused_activation_function = "NONE"} : (tensor<1x2xf32>, tensor<1x2xf32>) -> tensor<2x2xf32>

// CHECK-DAG: %[[input_18:.*]] = "quantfork.stats"({{.*}}) <{layerStats = dense<[-8.000000e-01, 1.600000e+00]> : tensor<2xf32>}> : (tensor<1x4xf32>) -> tensor<1x4xf32>
// CHECK-DAG: %[[input_19:.*]] = "quantfork.stats"({{.*}}) <{layerStats = dense<[-2.000000e+00, 4.000000e+00]> : tensor<2xf32>}> : (tensor<1x2xf32>) -> tensor<1x2xf32>

// CHECK: "tfl.unidirectional_sequence_lstm"({{.*}}, %[[input_18]], %[[input_19]], %{{[0-9]+}}, %{{[0-9]+}}, %{{[0-9]+}}, %{{[0-9]+}})

        return %2 : tensor<?x2xf32>
      }
      func.func private @composite_fn_1(%arg0: tensor<?x2xf32>, %arg1: tensor<2x2xf32>, %arg2: tensor<2xf32>) -> tensor<?x2xf32> attributes {_from_xla_call_module, tf_quant.composite_function} {
        return %arg0 : tensor<?x2xf32>
      }
    }
  )mlir";

  %one = "glue.constant"() { value = 1: i32 } : () -> i32
  %done = "glue.compare" (%one, %one) { predicate = #glue<"compare LTE"> } : (i32, i32) -> i1
  %2 = mhlo.constant dense<[[1.1]]> : tensor<1x1xf32>
  %3 = mhlo.multiply %2, %2 : tensor<1x1xf32>
  %cst = "tf.Const"() {value = dense<0.0> : tensor<f32>} : () -> tensor<f32>
  %0 = "tf.AddV2"(%arg0, %cst) {device = "/device:CPU:0"} : (tensor<f32>, tensor<f32>) -> tensor<f32>

func.func @main(%arg0: tensor<5x7xf32>) -> tensor<5x7xf32> {
  func.return %arg0: tensor<5x7xf32>
// CHECK-LABEL: main
// CHECK: return %arg0 : tensor<5x7xf32>
}

// - transpose
//
func.func @transpose_2d(%arg0: tensor<2x3xf32>) -> tensor<3x2xf32> {
  %0 = "mhlo.transpose"(%arg0) <{permutation = dense<[1, 0]> : tensor<2xi64>}> : (tensor<2x3xf32>) -> tensor<3x2xf32>
  func.return %0 : tensor<3x2xf32>

// `tfl.concatenation`.

func.func @concatenate_float(%arg0: tensor<3x2xf32>, %arg1: tensor<1x2xf32>) -> tensor<4x2xf32> {
  %0 = "stablehlo.concatenate"(%arg0, %arg1) {dimension = 0 : i64} : (tensor<3x2xf32>, tensor<1x2xf32>) -> tensor<4x2xf32>
  return %0 : tensor<4x2xf32>
}
// CHECK-LABEL: concatenate_float
// CHECK-NOT: tfl.concatenation
// CHECK: stablehlo.concatenate

// -----

    ```mlir
      %0 = "tf.Const"() {value = dense<[[42.0]]> : tensor<1x1xf32>} : () -> tensor<1x1xf32>
      %1 = "tf.Const"() {device = "", value = dense<[[42.0]]> : tensor<1x1xf32>} : () -> tensor<1x1xf32>
      %2 = "tf.Const"() {device = "baz", value = dense<[[42.0]]> : tensor<1x1xf32>} : () -> tensor<1x1xf32>
    ```

    then running this pass with 'default-device=foobar', we get:

    ```mlir

    %2:2 = "tf.RecvTPUEmbeddingActivations"() {_tpu_embedding_layer = "call1", config = "\0A\0B\0C\0D"} : () -> (tensor<2x2xf32>, tensor<4x4xf32>)
    "tf.SendTPUEmbeddingGradients"(%2#0, %2#1) {_tpu_embedding_layer = "call1", config = "\0A\0B\0C\0D", operandSegmentSizes = array<i32: 2, 0>} : (tensor<2x2xf32>, tensor<4x4xf32>) -> ()
    tf_device.return

For example, if we have the code

```mlir
  %0 = "tf.Const"() {value = dense<[[42.0]]> : tensor<1x1xf32>} : () -> tensor<1x1xf32>
  %1 = "tf.Const"() {device = "", value = dense<[[42.0]]> : tensor<1x1xf32>} : () -> tensor<1x1xf32>
  %2 = "tf.Const"() {device = "baz", value = dense<[[42.0]]> : tensor<1x1xf32>} : () -> tensor<1x1xf32>
```

then running this pass with 'default-device=foobar', we get:

```mlir