// should never reach here!
      throw new UnsupportedOperationException();
    }

    @Override
    public Entry<K, V> belowSamplesGreater() {
      // should never reach here!
      throw new UnsupportedOperationException();
    }

    @Override
    public Entry<K, V> aboveSamplesLesser() {
      // should never reach here!
      throw new UnsupportedOperationException();

    return x > 0 & (x & (x - 1)) == 0;
  }

  /**
   * Returns 1 if {@code x < y} as unsigned integers, and 0 otherwise. Assumes that x - y fits into
   * a signed int. The implementation is branch-free, and benchmarks suggest it is measurably (if
   * narrowly) faster than the straightforward ternary expression.
   */
  @VisibleForTesting
  static int lessThanBranchFree(int x, int y) {

  @CheckForNull
  public abstract C previous(C value);

  /**
   * Returns a signed value indicating how many nested invocations of {@link #next} (if positive) or
   * {@link #previous} (if negative) are needed to reach {@code end} starting from {@code start}.
   * For example, if {@code end = next(next(next(start)))}, then {@code distance(start, end) == 3}
   * and {@code distance(end, start) == -3}. As well, {@code distance(a, a)} is always zero.

   * than CPUs, supposing that each thread were bound to a CPU,
   * there would exist a perfect hash function mapping threads to
   * slots that eliminates collisions. When we reach capacity, we
   * search for this mapping by randomly varying the hash codes of
   * colliding threads.  Because search is random, and collisions
   * only become known via CAS failures, convergence can be slow,

            V entryValue = e.getValue();

            if (entryValue != null) {
              // TODO(kak): Remove this branch
            } else if (isCollected(e)) {
              // TODO(kak): Remove this branch
            } else {
              return null;
            }

            ++modCount;
            E newFirst = removeFromChain(first, e);

   *
   * <p>The graph below would be valid input with start nodes of {@code a, f, c}. However, if {@code
   * b} were <i>also</i> a start node, then there would be multiple paths to reach {@code e} and
   * {@code h}.
   *
   * <pre>{@code
   *    a     b      c
   *   / \   / \     |
   *  /   \ /   \    |
   * d     e     f   g
   *       |
   *       |
   *       h

    int initialBufferSize = min(BUFFER_SIZE, max(128, Integer.highestOneBit(totalLen) * 2));
    // Starting with an 8k buffer, double the size of each successive buffer. Smaller buffers
    // quadruple in size until they reach 8k, to minimize the number of small reads for longer
    // streams. Buffers are retained in a deque so that there's no copying between buffers while

      int timeout,
      TimeUnit unit)
      throws ExecutionException, TimeoutException {
    /*
     * For faster tests (that may hang indefinitely if the class under test has
     * a bug!), switch the second branch to call untimed future.get() instead of
     * pseudoTimedGet.
     */
    return (inputs.hasDelayed(iFuture, jFuture))
        ? pseudoTimedGetUninterruptibly(future, timeout, unit)

  @CheckForNull
  public abstract C previous(C value);

  /**
   * Returns a signed value indicating how many nested invocations of {@link #next} (if positive) or
   * {@link #previous} (if negative) are needed to reach {@code end} starting from {@code start}.
   * For example, if {@code end = next(next(next(start)))}, then {@code distance(start, end) == 3}
   * and {@code distance(end, start) == -3}. As well, {@code distance(a, a)} is always zero.

    return x > 0 & (x & (x - 1)) == 0;
  }

  /**
   * Returns 1 if {@code x < y} as unsigned longs, and 0 otherwise. Assumes that x - y fits into a
   * signed long. The implementation is branch-free, and benchmarks suggest it is measurably faster
   * than the straightforward ternary expression.
   */
  @VisibleForTesting
  static int lessThanBranchFree(long x, long y) {
    // Returns the sign bit of x - y.