%s1 = "tf.Const"() {value = dense<[1, 2]> : tensor<2xi32>} : () -> tensor<2xi32>
  %r0, %r1 = "tf.BroadcastGradientArgs"(%s0, %s1) {} : (tensor<2xi32>, tensor<2xi32>) -> (tensor<0xi32>, tensor<0xi32>)

  // CHECK-DAG: %[[R:.*]] = "tf.Const"() <{value = dense<> : tensor<0xi32>}> : () -> tensor<0xi32>
  // CHECK-NOT: tf.BroadcastGradientArgs
  // CHECK: return %[[R]], %[[R]]

  func.return %r0, %r1 : tensor<0xi32>, tensor<0xi32>
}

}

// -----

// CHECK-LABEL: func @testBroadcastGradientArgs
func.func @testBroadcastGradientArgs(%s0: tensor<4xi32>, %s1: tensor<4xi32>) -> (tensor<1xi32>, tensor<0xi32>) {
  // CHECK:     tf.BroadcastGradientArgs
  %r0, %r1 = "tf.BroadcastGradientArgs"(%s0, %s1) : (tensor<4xi32>, tensor<4xi32>) -> (tensor<1xi32>, tensor<0xi32>)
  func.return %r0, %r1 : tensor<1xi32>, tensor<0xi32>
}

// -----

// CHECK-LABEL: @acos

    %s1 = "tf.Const"() {value = dense<[  1, 1,  1, 1280]> : tensor<4xi32>} : () -> tensor<4xi32>

    // CHECK: "tf.BroadcastGradientArgs"
    // CHECK-NOT: _xla_outside_compilation
    %r0, %r1 = "tf.BroadcastGradientArgs"(%s0, %s1) {} : (tensor<4xi32>, tensor<4xi32>) -> (tensor<1xi32>, tensor<3xi32>)
    tf_device.return

  }

  // CHECK-LABEL: func @testBroadcastGradientArgs
  func.func @testBroadcastGradientArgs(%s0: tensor<4xi32>, %s1: tensor<4xi32>) -> (tensor<1xi32>, tensor<0xi32>) {
    // CHECK:     tf.BroadcastGradientArgs
    %r0, %r1 = "tf.BroadcastGradientArgs"(%s0, %s1) : (tensor<4xi32>, tensor<4xi32>) -> (tensor<1xi32>, tensor<0xi32>)
    func.return %r0, %r1 : tensor<1xi32>, tensor<0xi32>
  }

  // CHECK-LABEL: @acos

  func.func @op_non_fully_defined_shape_fn(%arg0: tensor<*xi32>, %arg1: tensor<0xi32>) -> tensor<?xi32> {
    // CHECK: tf.BroadcastGradientArgs
    // CHECK-SAME: (tensor<*xi32>, tensor<0xi32>) -> (tensor<?xi32>, tensor<?xi32>)
    %2:2 = "tf.BroadcastGradientArgs"(%arg0, %arg1) {T = "tfdtype$DT_INT32", name = "BroadcastGradientArgs"} : (tensor<*xi32>, tensor<0xi32>) -> (tensor<?xi32>, tensor<?xi32>)
    func.return %2#0 : tensor<?xi32>
  }

      "Range", "Rank", "Reshape", "Shape", "ShapeN", "Size", "Squeeze",
      "Transpose", "ZerosLike", "OnesLike", "BiasAdd" /*PW + Broadcast*/,
      "BroadcastArgs", "BroadcastGradientArgs", "OneHot", "Concat", "ConcatV2",
      "ConcatOffset", "Const", "MirrorPad", "MirrorPadGrad", "Pack", "Pad",
      "PadV2", "Reverse", "ReverseV2", "ReverseSequence", "Slice", "Split",

  SmallVector<int64_t, 4> s0_shape, s1_shape;
  DenseIntElementsAttr s0, s1;
  if (!ExtractInputConstShape(*this, s0, s1, s0_shape, s1_shape))
    return failure();

  // Fold BroadcastGradientArgs into two constants if both of the inputs have
  // known shape.
  SmallVector<int64_t, 4> bcasted_shape;
  // Verifier should already ensure the broadcast compatibility.
  bool bcast_compatible =

  );

  let results = (outs
    TF_I32OrI64Tensor:$r0
  );

  TF_DerivedOperandTypeAttr T = TF_DerivedOperandTypeAttr<0>;
}

def TF_BroadcastGradientArgsOp : TF_Op<"BroadcastGradientArgs", [Pure, SameOperandsAndResultElementType, TF_OperandHasRank<0, 1>, TF_OperandHasRank<1, 1>, TF_ResultHasRank<0, 1>, TF_ResultHasRank<1, 1>]> {
  let summary = [{