>  I ran into a very strange effect when some Sun folks tried to benchmark
>  JRuby's multi-thread scalability. In short, adding more threads actually
>  caused the benchmarks to take longer.

>  The source of the problem (at least the source that, when fixed, allowed
>  normal thread scaling), was an increment, mask, and test of a static int
>  field. The code in question looked like this:

>  private static int count = 0;

>  public void pollEvents(ThreadContext context) {
>    if ((count++ & 0xFF) == 0) context.poll();
>  }

>  So the basic idea was that this would call poll() every 256 hits,
>  incrementing a counter all the while. My first attempt to improve
>  performance was to comment out the body of poll() in case it was causing
>  a threading bottleneck (it does some locking and such), but that had no
>  effect. Then, as a total shot in the dark, I commented out the entire
>  line above. Thread scaling went to normal.