Re: [PATCH] xen: sched_credit: filter node-affinity mask against online cpus

[Dario Faggioli <dario.faggioli@citrix.com>]

[-- Attachment #1.1: Type: text/plain, Size: 1903 bytes --]

On lun, 2013-09-16 at 15:00 +0100, George Dunlap wrote:
> On 13/09/13 17:09, Dario Faggioli wrote:
>
> > diff --git a/xen/common/sched_credit.c b/xen/common/sched_credit.c
> > index dbe6de6..d76cd88 100644
> > --- a/xen/common/sched_credit.c
> > +++ b/xen/common/sched_credit.c
>
> > @@ -627,10 +638,18 @@ _csched_cpu_pick(const struct scheduler *ops, struct vcpu *vc, bool_t commit)
> >       int balance_step;
> >   
> >       online = cpupool_scheduler_cpumask(vc->domain->cpupool);
> > +    cpumask_and(&cpus, vc->cpu_affinity, online);
> > +
> >       for_each_csched_balance_step( balance_step )
> >       {
> > +        /*
> > +         * We filter the node-affinity mask against
> > +         * [vc->cpu_affinity && online] here to avoid problems if all
> > +         * the cpus in in the node-affinity mask are offline (e.g.,
> > +         * because the domain moved to a different cpupool).
> > +         */
> 
> It's probably worth mentioning that the reason you do the cpumask_and 
> here and not in the other place this is called is that cpumask_cycle is 
> called after this one (which will choke on an empy mask), but not in the 
> other place it's called.
> 
Well, it is indeed what the comment says, or at leas what I wanted it to
say. :-)

I tied to concentrate on the actual issue, which is avoid picking an
offline cpu, independently on which specific ASSERT is the one that
triggers.

Anyway, I can try to make the comment more accurate and descriptive
though.

> Other than that, looks good -- thanks.
> 
Ok, I'll resend with an updated comment.

Thanks and Regards,
Dario

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)


[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel