Comments on the KSE option

[David Xu]

On Saturday 28 October 2006 12:36, Julian Elischer wrote:
> >> Julian
> >
> > As you are emphasizing fairness, I must say I don't believe fairness in
> > libpthread either,
>
> you mean you don't think it is a good idea or that you don't think it
> works? (sorry, I know that your english is way better than my
> chinese ;-)
>
I meant I don't think libpthread's userland scheduler + ksegrp in kernel
has implemented fairness between threads correctly.

> > I don't think writing a fairness scheduler is an
> > easy work, does kernel have made fairness for threads in same ksegrp,
> > so does libpthread's userland scheduler ?
>
> The kernel is only responsible for making sure that one ksegrp
> (usually a process in my original idea) is not unfair to another
> ksegrp.
> What happens within the ksegrp is not it's interest. And no it
> isn't an easy thing to do which is why I had hoped that some
> PhD student would have taken it up by now :-)
>
> > they don't, it can make threads
> > in same ksegrp misbehaviored, so what we have done is still process
> > scheduling fairness even  there is ksegrp in kernel, and now sacrificed
> > fairness between threads.
>
> once again, I'm not sure what you mean by that.
>