}

// 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>
  %w = arith.constant dense<[[0.0, 1.0], [1.0, 2.0]]> : tensor<2x2xf32>
  %b = arith.constant dense<[0.0, 2.1473647e6]> : tensor<2xf32>

    has_token_input_output = false, module = "", platforms = [],
    version = 5 : i64
  } : (tensor<1x2xf32>, tensor<2x3xf32>) -> tensor<1x3xf32>
  return %0 : tensor<1x3xf32>
}

// CHECK-LABEL: func.func @qdq_for_dot_general_weight_empty
// CHECK-SAME: (%[[ARG_0:.+]]: tensor<1x2xf32>) -> tensor<1x3xf32>
// CHECK: %[[CST:.+]] = "tf.Const"() <{value = dense<3.000000e-01> : tensor<2x3xf32>}> : () -> tensor<2x3xf32>

  %elem_2 = "tf._SecondOp"() : () -> tensor<16x1xf32>
  %tl_set_item2 = "tf.TensorListSetItem"(%tl, %one, %elem_2) : (tensor<!tf_type.variant<tensor<?x1xf32>>>, tensor<i32>, tensor<16x1xf32>) -> tensor<!tf_type.variant<tensor<?x1xf32>>>
  func.return
}

// CHECK: use_two_sep_ops_empty_tensor_list

   %3 = "quantfork.qcast"(%arg0) {volatile} : (tensor<1x2xf32>) -> tensor<1x2x!quant.uniform<i8:f32, 6.000000e-03:-128>>
   %4 = "quantfork.dcast"(%3) : (tensor<1x2x!quant.uniform<i8:f32, 6.000000e-03:-128>>) -> tensor<1x2xf32>
// expected-error @+2 {{Failed to find a valid entry function}}

    // expected-remark@+1 {{Cost: 1}}
    %data = "tf.Const"() { value = dense<0.1> : tensor<476x28xf32> } : () -> tensor<476x28xf32>
    // expected-remark@+1 {{Cost: 28}}
    %1 = "tf.SparseSegmentSum"(%data, %indices, %segment_ids) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<476x28xf32>, tensor<3xi64>, tensor<3xi64>) -> tensor<?x28xf32>
    // expected-remark@+1 {{Cost: 3}}

    %7 = call @uniform_quantize_0(%arg0, %0, %1) : (tensor<1x2xf32>, tensor<1x1xf32>, tensor<1x1xi8>) -> tensor<1x2xi8>
    %8 = stablehlo.convert %7 : (tensor<1x2xi8>) -> tensor<1x2xf32>
    %9 = stablehlo.convert %2 : (tensor<2x3xi8>) -> tensor<2x3xf32>
    %10 = stablehlo.dot_general %8, %9, contracting_dims = [1] x [0] : (tensor<1x2xf32>, tensor<2x3xf32>) -> tensor<1x3xf32>

  %2:2 = "tf.IteratorGetNext"(%arg0) {device = "/device:CPU:0"} : (tensor<*x!tf_type.resource>) -> (tensor<128xf32>, tensor<128xf32>)
  %3:2 = "tf.IteratorGetNext"(%arg1) {device = "/device:CPU:0"} : (tensor<*x!tf_type.resource>) -> (tensor<128xf32>, tensor<128xf32>)
  // CHECK: "tf.TPUCompileSucceededAssert"(%[[COMPILE]]#0)
  "tf_device.launch"() ({

// 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]+}})

}

// 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>

  %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>)