XLA not working

hello, everyone
As the left sidebar shows, the tensorboard 1.14 .0 has supported display a graph with XLA enabled, but in my XLA test, the graph in tensorboard was not converted as i wish: the nodes added to the same XLA cluster would be replaced by a XlaCompileOp + XlaRunOp. At present, i can only figure out which node is compiled by its attribute, which way i still doubted because it seemed that the function related to XLA does not work well,
the source code i tested:

mnist_with_summaries.py
to enable XLA, i turn JIT by modifying the function:

  def nn_layer(input_tensor, input_dim, output_dim, layer_name, act=tf.nn.relu):
    """Reusable code for making a simple neural net layer.
    It does a matrix multiply, bias add, and then uses ReLU to nonlinearize.
    It also sets up name scoping so that the resultant graph is easy to read,
    and adds a number of summary ops.
    """
    jit_scope = tf.contrib.compiler.jit.experimental_jit_scope(compile_ops=True)
    with jit_scope:
      # Adding a name scope ensures logical grouping of the layers in the graph.
      with tf.name_scope(layer_name):
        # This Variable will hold the state of the weights for the layer
        with tf.name_scope('weights'):
          weights = weight_variable([input_dim, output_dim])
          variable_summaries(weights)
        with tf.name_scope('biases'):
          biases = bias_variable([output_dim])
          variable_summaries(biases)
        with tf.name_scope('Wx_plus_b'):
          preactivate = tf.matmul(input_tensor, weights) + biases
          tf.summary.histogram('pre_activations', preactivate)
      activations = act(preactivate, name='activation')
      tf.summary.histogram('activations', activations)
      return activations

the following is the original graph when the "Default" tag is selected:

but when i chose the after several steps, the graph made me confused.

Question 1: the "XLA Cluster" was still unselectable even though XLA was enabled
Question 2: Intuitively, all the converted to brown(unused structure) were added to a XLA Cluster, but i could not know that according to the attribute of the selected node. So, why this node changed to unused?
Question 3: just as question 2，the node in the middle graph, as i selected, clearly belonged to a XLA Cluster

In a word, i am confused by the current XLA function in Tensorboard. if these questions were caused by my unfamiliarity of tensorboard , thank you very much for giving me some help.