RE: Thread/VMThread proposal

[Jeroen Frijters]

Brian Jones wrote:
> You may wish to check that a "stillborn" thread has a null thread
> group after stop() has been called.  I did confirm already that
> threads that run and are stopped are seen to have a null thread group.

The Sun JDK ignores stop() when called on an unstarted thread (and the
group is not set to null), but this behavior is incorrect. It appears
that the only way to remove a thread from its group is to actually start
it and wait for it to die (which it is supposed to do immediately when
stop() was previously called).

[join]
> I think you can avoid polling/sleep.

Yes, I wasn't quite awake yet. I changed it to a wait/notifyAll solution
that is simple and elegant.

Regards,
Jeroen