Scott Stirling writes:
> Java is inherently
> multi-threaded and when using native threads (a must on multi-CPUs and on
> production JVMs from Sun) it inherits the threading model of the OS it runs
> on.
This is sometimes a disadvantage.
When writing heavy-duty multi-threaded apps for Java, you quite often
have to program very defensively, because, after all, you're writing
in Java, and you're striving to be portable. Since you don't know
{a priori} what the threading model is going to be like, you have to
code very carefully.
But, all in all, Java is a great language, and I'm just nit-picking.
> The savior may be something like
> Ralph Engelschall's GNU-threads. If the user-level threads can be
> implemented without modifying the kernel (or burdening the kernel developers
> too much), then it might happen.
It doesn't seem very likely that a user-level thread library will be
able to take advantage of a SMP machine, and that's a big strike
against this scheme.
--kevin
--
Co-worker: ``Well, I think I had a threading problem with my app, but
I solved it by calling Thread.yield()!!!''
Me: <shudder>
**********************************************************
To unsubscribe from this list, send mail to
[EMAIL PROTECTED] with the following text in the
*body* (*not* the subject line) of the letter:
unsubscribe gnhlug
**********************************************************
