* that it is not "API-frozen."
 *
 * <p>It is generally safe for <i>applications</i> to depend on beta APIs, at the cost of some extra
 * work during upgrades. However it is generally inadvisable for <i>libraries</i> (which get
 * included on users' CLASSPATHs, outside the library developers' control) to do so.
 *
 * @author Kevin Bourrillion
 */
@Retention(RetentionPolicy.CLASS)

    for details.

    The summary is that it ignores one file in favor of the other.
    util.concurrent, like nearly all our packages, has two .gwt.xml files: one
    for prod and one for tests. However, unlike our other packages, as of this
    writing it has test supersource but no prod supersource.

    GWT happens to use the prod .gwt.xml, so it looks for no supersource for

    for details.

    The summary is that it ignores one file in favor of the other.
    util.concurrent, like nearly all our packages, has two .gwt.xml files: one
    for prod and one for tests. However, unlike our other packages, as of this
    writing it has test supersource but no prod supersource.

    GWT happens to use the prod .gwt.xml, so it looks for no supersource for

   * <p>This is guaranteed to return zero if the dataset contains only exactly one finite value. It
   * is not guaranteed to return zero when the dataset consists of the same value multiple times,
   * due to numerical errors. However, it is guaranteed never to return a negative result.
   *
   * <h3>Non-finite values</h3>
   *
   * <p>If the dataset contains any non-finite values ({@link Double#POSITIVE_INFINITY}, {@link

       *
       * My impression is that an AtomicReferenceFieldUpdater in a static field is similarly fast to
       * Unsafe on modern JVMs (if perhaps not quite as fast as VarHandle?). However, I'm not sure
       * exactly what we've benchmarked, and we certainly haven't benchmarked as far back as JDK 8.
       * (We also haven't benchmarked under Android. We continue to use UnsafeAtomicHelper there so

        // Since SamrOpenAlias's constructor parameters are not directly exposed via getters in MsrpcSamrOpenAlias,
        // we cannot directly verify them here without reflection or extending SamrOpenAlias for testing.
        // However, the primary responsibility of MsrpcSamrOpenAlias's constructor is to call the super constructor
        // and set its own specific fields (ptype, flags).

    for details.

    The summary is that it ignores one file in favor of the other.
    util.concurrent, like nearly all our packages, has two .gwt.xml files: one
    for prod and one for tests. However, unlike our other packages, as of this
    writing it has test supersource but no prod supersource.

    GWT happens to use the prod .gwt.xml, so it looks for no supersource for

 * }
 * }
 *
 * @author Ben Yu
 * @since 12.0
 */
/*
 * A nullable bound would let users create a TypeParameter instance for a parameter with a nullable
 * bound. However, it would also let them create `new TypeParameter<@Nullable T>() {}`, which
 * wouldn't behave as users might expect. Additionally, it's not clear how the TypeToken API could

                    if (NbtAddress.getWINSAddress() == null) {
                        if (LogStream.level > 1) {
                            log.println("NetBIOS resolveOrder specifies WINS however the "
                                    + "jcifs.smb1.netbios.wins property has not been set");
                        }
                        continue;
                    }

   * calling code written in Kotlin).) Typically, we want to let a {@link Throwable} from such a
   * method propagate untouched, just as we'd typically let it do for a non-reflective call.
   * However, we can't usually write {@code throw t;} when {@code t} has a static type of {@link
   * Throwable}. But we <i>can</i> write {@code sneakyThrow(t);}.
   *