}

// 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>

      %2 = "tf.ReadVariableOp"(%arg15) : (tensor<*x!tf_type.resource<tensor<7x7x3x64xf32>>>) -> tensor<7x7x3x64xf32>
      %3 = "tf.ReadVariableOp"(%arg16) : (tensor<*x!tf_type.resource<tensor<1001xf32>>>) -> tensor<1001xf32>
      %4 = "tf.ReadVariableOp"(%arg17) : (tensor<*x!tf_type.resource<tensor<64x1001xf32>>>) -> tensor<64x1001xf32>
      %5 = "tf.ReadVariableOp"(%arg18) : (tensor<*x!tf_type.resource<tensor<f32>>>) -> tensor<f32>

  %0 = "tf_device.cluster_func"(%arg0, %arg1) {func = @_func, use_spmd_for_xla_partitioning = true, use_tpu = true, num_cores_per_replica = 2 : i64} : (tensor<2x4xf32>, tensor<4x2xf32>) -> tensor<2x2xf32>
  %1:2 = "tf.TPUPartitionedOutputV2"(%0) {device = "", partition_dims = [2, 1]} : (tensor<2x2xf32>) -> (tensor<1x2xf32>, tensor<1x2xf32>)

func.func @float_matmul(
  %arg0: tensor<1x10xf32>, %arg1: tensor<10x10xf32>) -> (tensor<*xf32>, tensor<*xf32>, tensor<*xf32>) {
  %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>

      %arg0 = "tf.Const"() {value = dense<-3.0> : tensor<5x7xf32>} : () -> tensor<5x7xf32>
      %arg1 = "tf.Const"() {value = dense<-3.0> : tensor<11x7xf32>} : () -> tensor<11x7xf32>

      %1 = "tf.MatMul"(%arg0, %arg1) {transpose_a = false, transpose_b = true} : (tensor<5x7xf32>, tensor<11x7xf32>) -> tensor<5x11xf32>

      func.return %1 : tensor<5x11xf32>
    }
  })";
  TF_ASSERT_OK_AND_ASSIGN(

    "vhlo.return_v1"(%lhs) : (tensor<f32>) -> ()
  }): (tensor<200x100x300xf32>, tensor<10x2xi32>, tensor<10x300xf32>) ->
      tensor<200x100x300xf32>
  func.return %0 : tensor<200x100x300xf32>
}

// CHECK-LABEL: func.func private @scatter(%arg0: tensor<200x100x300xf32>, %arg1: tensor<10x2xi32>, %arg2: tensor<10x300xf32>) -> tensor<200x100x300xf32> {

  constexpr absl::string_view kProgram = R"mlir(
func.func @main(
    %arg0: tensor<10x10xf32>, %scale: tensor<10xf32>, %zp: tensor<10xi32>
  ) -> tensor<10x10xi8> {
  %0 = "tf.UniformQuantize"(%arg0, %scale, %zp) {
    quantization_axis = 1 : i64,
    quantization_min_val = -128 : i64,
    quantization_max_val = 127 : i64
  } : (tensor<10x10xf32>, tensor<10xf32>, tensor<10xi32>) -> tensor<10x10x!tf_type.qint8>

  func.return %0 : tensor<i32>
}

// -----

// CHECK-LABEL: constant_tensor_vector
func.func @constant_tensor_vector(%arg0: vector<1x2xi32>) -> tensor<1x2xi32> {
  %0 = "tfr.constant_tensor"(%arg0) : (vector<1x2xi32>) -> tensor<1x2xi32>
  func.return %0 : tensor<1x2xi32>
}

// -----

// CHECK-LABEL: constant_tensor_array
func.func @constant_tensor_array() -> !tfr.tensor {

// Tests for ops with canonicalization patterns.

// CHECK-LABEL: get_real_shape
func.func @get_real_shape(%arg0: tensor<1x2xf32>) -> tensor<2xindex> {
  %0 = "tfr.cast"(%arg0) : (tensor<1x2xf32>) -> !tfr.tensor
  %1 = tfr.get_shape %0 -> !shape.shape
  %2 = shape.to_extent_tensor %1 : !shape.shape -> tensor<2xindex>
  func.return %2 : tensor<2xindex>

      %2 = "tf.XlaCallModule"(%arg0, %1, %0) <{Sout = [#tf_type.shape<?x2>], module = "", version = 9 : i64}> {_original_entry_function = "composite_fn_1"} : (tensor<?x2xf32>, tensor<2x2xf32>, tensor<2xf32>) -> tensor<?x2xf32>
      return %2 : tensor<?x2xf32>
    }