// This limits the number of pushes that can happen concurrently
			semaphore <- struct{}{}

			// Get the next proxy to push. This will block if there are no updates required.
			client, push, shuttingdown := queue.Dequeue()
			if shuttingdown {
				return
			}
			recordPushTriggers(push.Reason)
			// Signals that a push is done by reading from the semaphore, allowing another send on it.
			doneFunc := func() {

	}

	var serverShuttingDownCh <-chan struct{}
	if signals := apirequest.ServerShutdownSignalFrom(req.Context()); signals != nil {
		serverShuttingDownCh = signals.ShuttingDown()
	}

	server := &WatchServer{
		Watching: watcher,
		Scope:    scope,

		UseTextFraming:  useTextFraming,
		MediaType:       mediaType,
		Framer:          framer,
		Encoder:         encoder,

		// this simulates a watch well enough for our test
		signals := apirequest.ServerShutdownSignalFrom(req.Context())
		if signals == nil {
			w.WriteHeader(http.StatusInternalServerError)
			return
		}
		<-signals.ShuttingDown()
		w.WriteHeader(http.StatusOK)
	})
	s.Handler.NonGoRestfulMux.Handle("/apis/watches.group/v1/namespaces/foo/bar", handler)
	return inflight
}

      TF_EXCLUSIVE_LOCKS_REQUIRED(execution_mutex_);

  enum class ExecutionState {
    kReadyToExecute,
    kHasResult,
    kIdle,
    kShuttingDown,
  };

  tensorflow::mutex execution_mutex_;
  ExecutionState execution_state_ TF_GUARDED_BY(execution_mutex_) =
      ExecutionState::kIdle;
  // Tells the worker thread that there is new work.