* weak references.
 *
 * <p><strong>Performance</strong>
 *
 * <p>The extra bookkeeping done by cycle detecting locks comes at some cost to performance.
 * Benchmarks (as of December 2011) show that:
 *
 * <ul>
 *   <li>for an unnested {@code lock()} and {@code unlock()}, a cycle detecting lock takes 38ns as
 *       opposed to the 24ns taken by a plain lock.

 *   }
 * } else {
 *   // do other things since the monitor was not available
 * }
 * }</pre>
 *
 * <h2>Comparison with {@code synchronized} and {@code ReentrantLock}</h2>
 *
 * <p>The following examples show a simple threadsafe holder expressed using {@code synchronized},
 * {@link ReentrantLock}, and {@code Monitor}.
 *
 * <h3>{@code synchronized}</h3>
 *

    successfulAsList(ImmutableList.of(SettableFuture.create())).cancel(true);

    /*
     * Complex (combined.cancel -> input.cancel -> other.set -> setOneValue),
     * to show that this isn't just about problems with the input future we just
     * cancelled:
     */
    final SettableFuture<String> future1 = SettableFuture.create();

    public Container(T data) {}
  }

  public <T extends Container<String>> void testConstructor_parameterTypes()
      throws NoSuchMethodException {
    @SuppressWarnings("rawtypes") // Reflection API skew
    Constructor<Container> constructor = Container.class.getConstructor(Object.class);
    Invokable<T, ?> invokable = new TypeToken<T>(getClass()) {}.constructor(constructor);

    successfulAsList(ImmutableList.of(SettableFuture.create())).cancel(true);

    /*
     * Complex (combined.cancel -> input.cancel -> other.set -> setOneValue),
     * to show that this isn't just about problems with the input future we just
     * cancelled:
     */
    final SettableFuture<String> future1 = SettableFuture.create();

          <show>package</show>
        </configuration>
      </plugin>
      <!-- Disable "normal" testing, which doesn't work for GWT tests. -->
      <plugin>
        <artifactId>maven-surefire-plugin</artifactId>

     * constructor's. This is an arbitrary rule since no existing language spec mandates one way or
     * the other. From the declaration syntax, the class type parameter appears first, but the call
     * syntax may show up in opposite order such as {@code new <A>Foo<B>()}.
     */
    @Override
    public final TypeVariable<?>[] getTypeParameters() {
      TypeVariable<?>[] declaredByClass = getDeclaringClass().getTypeParameters();

    public Container(T data) {}
  }

  public <T extends Container<String>> void testConstructor_parameterTypes()
      throws NoSuchMethodException {
    @SuppressWarnings("rawtypes") // Reflection API skew
    Constructor<Container> constructor = Container.class.getConstructor(Object.class);
    Invokable<T, ?> invokable = new TypeToken<T>(getClass()) {}.constructor(constructor);