On Tue, 15 Nov 2016, Thomas Gleixner wrote:
> On Fri, 11 Nov 2016, Fenghua Yu wrote:
> > @@ -801,6 +801,7 @@ static void rmdir_all_sub(void)
> > {
> > struct rdtgroup *rdtgrp, *tmp;
> > struct task_struct *p, *t;
> > + int cpu;
> >
> > /* move all tasks to default resource group */
> > read_lock(&tasklist_lock);
> > @@ -819,10 +820,19 @@ static void rmdir_all_sub(void)
> > /* Remove each rdtgroup other than root */
> > if (rdtgrp == &rdtgroup_default)
> > continue;
> > +
> > + /* Give any CPUs back to the default group */
> > + cpumask_or(&rdtgroup_default.cpu_mask,
> > + &rdtgroup_default.cpu_mask, &rdtgrp->cpu_mask);
>
> That's a pointless exercise. We can simply copy cpu_online_mask to the
> default group mask and be done with it. I'll fix that up.
Actually we cannot. A CPU which already run the offline callback might
still be marked online. So the above is correct.
Thanks,
tglx
