// will take ownership and maintain devices' lifetime.
  virtual Status AddDevices(std::vector<std::unique_ptr<Device>> devices) = 0;

  // Block until all pending nodes are finished.
  virtual Status AsyncWait() = 0;

  // Add a function (serialized FunctionDef protocol buffer) so that it can
  // be executed as an op. Return error if the function with the same name
  // already exists.

  status->status = absl::OkStatus();
}

TFE_Executor* TFE_NewExecutor(bool is_async, bool enable_streaming_enqueue,
                              int in_flight_nodes_limit) {
  return new TFE_Executor(is_async, enable_streaming_enqueue,
                          in_flight_nodes_limit);
}

void TFE_DeleteExecutor(TFE_Executor* executor) { delete executor; }

bool TFE_ExecutorIsAsync(TFE_Executor* executor) {

*   `tf.saved_model.SaveOptions`

    * `SaveOptions` now takes a new argument called
      `experimental_debug_stripper`. When enabled, this strips the debug nodes
      from both the node defs and the function defs of the graph. Note that
      this currently only strips the `Assert` nodes from the graph and
      converts them into `NoOp`s instead.

*   `tf.data`

    * `tf.data` now has an `autotune_options.initial_parallelism` option to

abstract: TensorFlow is a machine learning system that operates at large scale and in heterogeneous environments. TensorFlow uses dataflow graphs to represent computation, shared state, and the operations that mutate that state. It maps the nodes of a dataflow graph across many machines in a cluster, and within a machine across multiple computational devices, including multicore CPUs, general purpose GPUs, and custom-designed ASICs known as Tensor Processing Units (TPUs). This architecture...

      // Name
      "AssertAndCollective",
      // Args
      {"x: float", "condition: bool"},
      // Return values
      {"y: float"},
      // Attr def
      {},
      // Nodes
      {
          {{"assert"},
           "Assert",
           {"condition", "x"},
           {{"T", std::vector<DataType>{DT_FLOAT}}}},
          {{"y"},
           "CollectiveReduce",
           {"x"},

  in_grads.reserve(input_tensors.size());
  for (int target_index = 0; target_index < input_tensors.size();
       ++target_index) {
    const auto current_grad =
        accumulated_gradients_.find(input_tensors[target_index].GetID());
    if (current_grad == accumulated_gradients_.end()) {
      if (IsDtypeTrainable(input_tensors[target_index].GetDType())) {

 
diff --git a/sysdeps/unix/sysv/linux/if_index.c b/sysdeps/unix/sysv/linux/if_index.c
index 8ba5eae..b620d21 100644
--- a/sysdeps/unix/sysv/linux/if_index.c
+++ b/sysdeps/unix/sysv/linux/if_index.c
@@ -38,12 +38,19 @@ __if_nametoindex (const char *ifname)
   return 0;
 #else
   struct ifreq ifr;
+  if (strlen (ifname) >= IFNAMSIZ)
+    {
+      __set_errno (ENODEV);
+      return 0;
+    }
+

    symlink_force(filepath, new_filepath)


def validate_cuda_config(environ_cp):
  """Run find_cuda_config.py and return cuda_toolkit_path, or None."""

  def maybe_encode_env(env):
    """Encodes unicode in env to str on Windows python 2.x."""
    if not is_windows() or sys.version_info[0] != 2:
      return env
    for k, v in env.items():
      if isinstance(k, unicode):
        k = k.encode('ascii')

    }
  }

  // Compute body nodes.
  std::vector<const Node*> body_nodes;
  status->status = tensorflow::ComputeBodyNodes(
      fn_body, fn_name, num_opers, opers, input_nodes, &body_nodes);
  if (TF_GetCode(status) != TF_OK) return nullptr;

  // Compute body nodes.
  std::vector<const Node*> control_output_nodes;
  control_output_nodes.reserve(ncontrol_outputs);

  for (int device_index = 0; device_index < underlying_devices_.size();
       ++device_index) {
    DeviceThread* device_thread = device_threads_[device_index].get();
    std::vector<TFE_TensorHandle*> device_inputs;
    device_inputs.reserve(inputs.size());
    for (int input_index = 0; input_index < inputs.size(); ++input_index) {
      // Parallel tensors are divided between operations by device.