*   `tf.saved_model.SaveOptions`

    *   Provided a new `experimental_skip_saver` argument which, if specified, will suppress the addition of `SavedModel`-native save and restore ops to the `SavedModel`, for cases where users already build custom save/restore ops and checkpoint formats for the model being saved, and the creation of the SavedModel-native save/restore ops simply cause longer model serialization times.

// In the TFv1 convention, TF_Tensor can hold a string serialization of
// DT_RESOURCE. The string serialization is converted back to a
// ResourceHandle during Session run where the TF_Tensor is converted to a
// Tensor.
// TFv2 does not depend on this conversion. There is no matching
// TF_TensorFromTensorV1 because the conversion to string is performed by the
// python side of Session.

  // threadpool, so that we avoid the possibility of running the runner_ in the
  // threadpool of GPU event mgr, as that can trigger more callbacks to be
  // scheduled on that same threadpool, causing a deadlock in cases where the
  // caller of event_mgr->ThenExecute() blocks on the completion of the callback
  // (as in the case of ConstOp kernel creation on GPU, which involves copying a
  // CPU tensor to GPU).