func.return %0: tensor<3x2xf16>
}

func.func @sparse_qu8() -> tensor<3x2x!quant.uniform<u8<1:255>:f32, 1.0>> {
  // CHECK-LABEL: @sparse_qu8

  func.return %0: tensor<3x2xf16>
}

func.func @sparse_qu8() -> tensor<3x2x!quant.uniform<u8<1:255>:f32, 1.0>> {
  // CHECK-LABEL: @sparse_qu8

  %3 = "tf.AddV2"(%rhs_dilation, %rhs_dilation) {device = ""} : (tensor<3xi32>, tensor<3xi32>) -> tensor<3xi32>
  %4 = "tf.AddV2"(%padding, %padding) {device = ""} : (tensor<3x2xi32>, tensor<3x2xi32>) -> tensor<3x2xi32>
  %5 = "tf.AddV2"(%strides, %strides) {device = ""} : (tensor<3xi32>, tensor<3xi32>) -> tensor<3xi32>

  return %0 : tensor<8x4x14x14x16xf32>
}

  // CHECK-SAME: -> tensor<3x2x!quant.uniform<i32:f32, 2.000000e+00:4>>
  // CHECK: %[[RES_INT:.*]] = mhlo.bitcast_convert %[[RES]] : (tensor<3x2x!quant.uniform<i32:f32, 2.000000e+00:4>>) -> tensor<3x2xi32>
  // CHECK: return %[[RES_INT]] : tensor<3x2xi32>

  %0 = "tf.UniformQuantizedAdd"(
    %arg0, %bias,
    %input_scales, %input_zps,
    %bias_scales, %bias_zps,
    %output_scales, %output_zps) {

func.func @equal_broadcast(%arg0: tensor<1xi32>, %arg1: tensor<1x2xi32>) -> tensor<1x2xi1> {
  // CHECK-DAG: %[[LHS_BCAST:.+]] = "mhlo.broadcast_in_dim"(%arg0) <{broadcast_dimensions = dense<1> : tensor<1xi64>}>
  // CHECK-NEXT: mhlo.compare EQ, %[[LHS_BCAST]], %arg1
  %0 = "tf.Equal"(%arg0, %arg1) : (tensor<1xi32>, tensor<1x2xi32>) -> tensor<1x2xi1>
  func.return %0: tensor<1x2xi1>
}

// CHECK-LABEL: func @equal_broadcast_no_incompatible_shapes_error

    // CHECK: %[[RESHAPE:.*]] = stablehlo.reshape %[[CALL]]

    // CHECK: %[[REDUCE:.*]] = "stablehlo.reduce_window"(%[[RESHAPE]], %[[Q0]])
    // CHECK{LITERAL}: padding = dense<[[0, 0], [1, 1], [0, 0]]> : tensor<3x2xi64>
    // CHECK-SAME: window_dimensions = array<i64: 1, 3, 1>
    // CHECK: %[[ARG1:.*]]: tensor<!quant.uniform<i8:f32, 3.000000e-01:1>>, %[[ARG2:.*]]: tensor<!quant.uniform<i8:f32, 3.000000e-01:1>>

  module attributes {tf.versions = {bad_consumers = [], min_consumer = 0 : i32, producer = 268 : i32}} {
  func.func @main() -> () {
    %cst0 = "tf.Const"(){ value = dense<0> : tensor<3x5xi1>} : () -> tensor<3x5xi1>
    %0 = "tf.Where"(%cst0) : (tensor<3x5xi1>) -> tensor<?x2xi64>
    func.return
  }
})";

absl::StatusOr<XlaCompiler::CompilationResult> CompileMlirModule(
    const char* mlir_module_str,

}

func.func @testPaddedDilatedConv(%arg0 : tensor<2x1920x64xf32>) ->  tensor<2x1920x128xf32> {
  %0 = "tf.Const"() {value = dense<[[0, 0], [2, 0], [0, 0]]> : tensor<3x2xi32>} : () -> tensor<3x2xi32>
  %1 = "tf.Const"() {value = dense<2> : tensor<1xi32>} : () -> tensor<1xi32>
  %2 = "tf.Const"() {value = dense<2> : tensor<i32>} : () -> tensor<i32>

    func.return %0: tensor<3x4xi32>
  }

  // CHECK-LABEL: @sparse_to_dense
  // CHECK-SAME: (%[[ARG0:.*]]: tensor<3x2xi32>, %[[ARG1:.*]]: tensor<3xf32>, %[[ARG2:.*]]: tensor<f32>)
  func.func @sparse_to_dense(%arg0: tensor<3x2xi32>, %arg1: tensor<3xf32>, %arg2: tensor<f32>) -> tensor<3x3xf32> {

func.func @transpose_2d_splat() -> tensor<3x2xi32> {
  %cst = arith.constant dense<0> : tensor<2x3xi32>
  %cst_perm = arith.constant dense<[1, 0]> : tensor<2xi32>

  // CHECK: %[[CST:.*]] = arith.constant dense<0> : tensor<3x2xi32>
  // CHECK: return %[[CST]]
  %0 = "tfl.transpose"(%cst, %cst_perm) : (tensor<2x3xi32>, tensor<2xi32>) -> tensor<3x2xi32>
  func.return %0 : tensor<3x2xi32>
}

// CHECK-LABEL: @transpose_2d_identity