Actually, this depends on whether the other threads got their reference
to the RubyThread through correctly synchronized means. So long as
they did, everything should be okay. But getting an NPE from an attempt
to call through threadImpl makes it sound like things aren't...
-mental
On Wed, 17 Oct 2007 9:20:47 -0700, MenTaLguY <[EMAIL PROTECTED]> wrote:
> On Wed, 17 Oct 2007 10:39:10 -0500, Charles Oliver Nutter
> <[EMAIL PROTECTED]> wrote:
>> Well now that is unusual. I've never seen or expected it to be possible
>> to have Thread#setPriority throw a NPE. I'll poke around a bit on that
>> one.
>
> One issue is that threads which do not descend from the Java thread that
> initialized the runtime (or were created prior to the setup of the Ruby
> Thread class in that runtime) are not guaranteed to see the initialized
> value
> of RubyThread.threadImpl, since there is no other synchronization of reads
> on it.
>
> I'd actually be a little surprised if that was the problem here, but
> it seems like something that should be addressed at some point, probably
> by making threadImpl a final field.
>
> -mental
>
>
> -------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe from this list please visit:
http://xircles.codehaus.org/manage_email
