/**
   * When we need a new thread to run tasks, we call [Backend.execute]. A few microseconds later we
   * expect a newly-started thread to call [Runnable.run]. We shouldn't request new threads until
   * the already-requested ones are in service, otherwise we might create more threads than we need.
   *
   * We use [executeCallCount] and [runCallCount] to defend against starting more threads than we

	if os.Getenv("GO_EXEC_TEST_PID") == "" {
		return
	}

	// Start some threads. 10 is arbitrary but intended to be enough
	// to ensure that the code won't have to create any threads itself.
	// In particular this should be more than the number of threads
	// the garbage collector might create.
	const threads = 10

	var wg sync.WaitGroup
	wg.Add(threads)
	ts := syscall.NsecToTimespec((100 * time.Microsecond).Nanoseconds())

#### Do-stuff-later pool

Sometimes there's an action required like calling the application layer or responding to a ping, and the thread discovering the action is not the thread that should do the work. We enqueue a runnable on this executor and it gets handled by one of the executor's threads.

### Locks

  }

  /**
   * Add a shutdown hook to wait for thread completion in the given {@link ExecutorService service}.
   * This is useful if the given service uses daemon threads, and we want to keep the JVM from
   * exiting immediately on shutdown, instead giving these daemon threads a chance to terminate
   * normally.
   *
   * @param service ExecutorService which uses daemon threads

   * running will have the given name.
   *
   * @param callable The callable to wrap
   * @param nameSupplier The supplier of thread names, {@link Supplier#get get} will be called once
   *     for each invocation of the wrapped callable.
   */
  @J2ktIncompatible
  @GwtIncompatible // threads
  static <T extends @Nullable Object> Callable<T> threadRenaming(

	// This function depends on several properties:
	//
	// 1. All OS threads that already exist are associated with an M in
	//    allm. i.e., we won't miss any pre-existing threads.
	// 2. All Ms listed in allm will eventually have an OS thread exist.
	//    i.e., they will set procid and be able to receive signals.
	// 3. OS threads created after we read allm will clone from a thread
	//    that has executed the system call. i.e., they inherit the

   * running will have the given name.
   *
   * @param callable The callable to wrap
   * @param nameSupplier The supplier of thread names, {@link Supplier#get get} will be called once
   *     for each invocation of the wrapped callable.
   */
  @J2ktIncompatible
  @GwtIncompatible // threads
  static <T extends @Nullable Object> Callable<T> threadRenaming(

          public @Nullable Void call() throws Exception {
            assertEquals(Thread.currentThread().getName(), newName.get());
            return null;
          }
        };
    Callables.threadRenaming(callable, newName).call();
    assertEquals(oldName, Thread.currentThread().getName());
  }

  @J2ktIncompatible
  @GwtIncompatible // threads
  public void testRenaming_exceptionalReturn() throws Exception {

          public @Nullable Void call() throws Exception {
            assertEquals(Thread.currentThread().getName(), newName.get());
            return null;
          }
        };
    Callables.threadRenaming(callable, newName).call();
    assertEquals(oldName, Thread.currentThread().getName());
  }

  @J2ktIncompatible
  @GwtIncompatible // threads
  public void testRenaming_exceptionalReturn() throws Exception {

 *
 * <p>Implementations are not guaranteed to be completely thread-safe.
 * However, the implementations:
 * <ul>
 * <li>should allow separate threads for dispatching and receiving, i.e. single thread that dispatches
 * and a different single thread that receives should be perfectly safe</li>
 * <li>should allow stopping or requesting stopping from a different thread than receiving/dispatching</li>