%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>
}

func.func @scatter_nd_add(%arg0: tensor<7xi64>, %arg1: tensor<1x1xi32>, %arg2: tensor<1xi64>) -> tensor<7xi64> {
  %0 = "tf.TensorScatterAdd"(%arg0, %arg1, %arg2) : (tensor<7xi64>, tensor<1x1xi32>, tensor<1xi64>) -> tensor<7xi64>
  func.return %0 : tensor<7xi64>

  // CHECK-LABEL: scatter_nd_add
  // CHECK:  %[[GATHER:.*]] = "tf.GatherNd"(%arg0, %arg1) <{bad_indices_policy = ""}> : (tensor<7xi64>, tensor<1x1xi32>) -> tensor<1xi64>

}

// CHECK-LABEL: func @testAdd() -> tensor<2x2xi32>
func.func @testAdd() -> tensor<2x2xi32> {
^bb0:
  %0 = arith.constant dense<[[0, 1], [2, 3]]> : tensor<2x2xi32>
  %1 = arith.constant dense<1> : tensor<2xi32>
  %2 = "tf.Add"(%0, %1) {device = "", name = "add"} : (tensor<2x2xi32>, tensor<2xi32>) -> tensor<2x2xi32>

          }
        }
        tensor_content: "\x01\x00\x00\x00\x03\x00\x00\x00\x02\x00\x00\x00\x05\x00\x00\x00\x04\x00\x00\x00\x07\x00\x00\x00"
        # CHECK: value = dense<{{\[\[}}1, 3, 2], [5, 4, 7]]> : tensor<2x3xi32>
      }
    }
  }

}

func.func @einsum_matmul(%arg0: tensor<7x9xf32>, %arg1: tensor<9x5xf32>) -> tensor<7x5xf32> {
  %0 = "tf.Einsum"(%arg0, %arg1) {T = "tfdtype$DT_FLOAT", equation = "ae,ed->ad"}: (tensor<7x9xf32>, tensor<9x5xf32>) -> tensor<7x5xf32>
  func.return %0 : tensor<7x5xf32>
  // CHECK-LABEL: einsum_matmul
  // CHECK: %[[v0:.*]] = "tf.BatchMatMulV2"(%arg0, %arg1) <{adj_x = false, adj_y = false}> : (tensor<7x9xf32>, tensor<9x5xf32>) -> tensor<7x5xf32>

        }) : () -> tensor<f32>
        "tf_device.cluster"() ({
          %group = "tf.Const"() <{value = dense<[[0, 1]]> : tensor<1x2xi32>}> : () -> tensor<1x2xi32>
          %arg1_reduced = "tf.XlaAllReduce"(%arg1_id, %group) <{mode = "CrossReplica", reduce_op = "Add"}> : (tensor<f32>, tensor<1x2xi32>) -> tensor<f32>
          "tf.OpA"(%arg1_reduced) : (tensor<f32>) -> ()
          tf_device.return
        }) : () -> ()

// Note: Argument quantization sequence omitted.
// CHECK: stablehlo.dot_general %{{.+}}, %{{.+}}, contracting_dims = [1] x [0] : (tensor<1x1024xi8>, tensor<1024x3xi8>) -> tensor<1x3xi32>

// Note: Result dequantization sequence omitted.
// CHECK: return %{{.+}} : tensor<1x3xf32>

// -----

// Tests that a simple dot_general (lifted as a function) with CustomAggregators

  %0 = "tfl.broadcast_to"(%arg0, %arg1) : (tensor<3xi32>, tensor<2xi32>) -> tensor<3x3xi32>
  return %0 : tensor<3x3xi32>
  // CHECK:  %cst = arith.constant dense<1> : tensor<3x3xi32>
  // CHECK:  %0 = tfl.mul(%arg0, %cst) <{fused_activation_function = "NONE"}> : (tensor<3xi32>, tensor<3x3xi32>) -> tensor<3x3xi32>
  // CHECK:  return %0 : tensor<3x3xi32>
}

  %1 = "tf.Concat"(%0, %arg0, %arg1) {device = "/job:localhost/replica:0/task:0/device:GPU:0"} : (tensor<i32>, tensor<2x1xi32>, tensor<2x1xi32>) -> tensor<2x2xi32>
  func.return %1 : tensor<2x2xi32>
}

// CHECK-LABEL: testConcatCwiseUnary
func.func @testConcatCwiseUnary(%arg0: tensor<?x1xf32>, %arg1: tensor<?x1xf32>, %arg2: tensor<i32>) -> tensor<?x2xf32> {

    %0 = "tf.XlaDynamicUpdateSlice"(%arg0, %arg1, %arg2) : (tensor<3x4xi32>, tensor<2x2xi32>, tensor<2xi32>) -> tensor<3x4xi32>
    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> {