On Fri, Feb 03, 2023 at 11:00:24AM -1000, Tejun Heo wrote:
> On Fri, Feb 03, 2023 at 11:40:40AM -0500, Waiman Long wrote:
> > Since commit 8f9ea86fdf99 ("sched: Always preserve the user
> > requested cpumask"), relax_compatible_cpus_allowed_ptr() is calling
> > __sched_setaffinity() unconditionally. This helps to expose a bug in
> > the current cpuset hotplug code where the cpumasks of the tasks in
> > the top cpuset are not updated at all when some CPUs become online or
> > offline. It is likely caused by the fact that some of the tasks in the
> > top cpuset, like percpu kthreads, cannot have their cpu affinity changed.
> >
> > One way to reproduce this as suggested by Peter is:
> > - boot machine
> > - offline all CPUs except one
> > - taskset -p ffffffff $$
> > - online all CPUs
> >
> > Fix this by allowing cpuset_cpus_allowed() to return a wider mask that
> > includes offline CPUs for those tasks that are in the top cpuset. For
> > tasks not in the top cpuset, the old rule applies and only online CPUs
> > will be returned in the mask since hotplug events will update their
> > cpumasks accordingly.
> >
> > Fixes: 8f9ea86fdf99 ("sched: Always preserve the user requested cpumask")
> > Reported-by: Will Deacon <will@xxxxxxxxxx>
> > Originally-from: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
> > Signed-off-by: Waiman Long <longman@xxxxxxxxxx>
>
> So, this is the replacement for the first patch[1] Will posted, right?
>
> > void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask)
> > {
> > unsigned long flags;
> > + struct cpuset *cs;
> >
> > spin_lock_irqsave(&callback_lock, flags);
> > - guarantee_online_cpus(tsk, pmask);
> > + rcu_read_lock();
> > +
> > + cs = task_cs(tsk);
> > + if (cs != &top_cpuset)
> > + guarantee_online_cpus(tsk, pmask);
> > + /*
> > + * TODO: Tasks in the top cpuset won't get update to their cpumasks
> > + * when a hotplug online/offline event happens. So we include all
> > + * offline cpus in the allowed cpu list.
> > + */
> > + if ((cs == &top_cpuset) || cpumask_empty(pmask)) {
> > + const struct cpumask *possible_mask = task_cpu_possible_mask(tsk);
> > +
> > + /*
> > + * We first exclude cpus allocated to partitions. If there is no
> > + * allowable online cpu left, we fall back to all possible cpus.
> > + */
> > + cpumask_andnot(pmask, possible_mask, top_cpuset.subparts_cpus);
>
> and the differences are that
>
> * It's only applied to the root cgroup.
>
> * Cpus taken up by partitions are excluded.
>
> Is my understanding correct?
>
> > + if (!cpumask_intersects(pmask, cpu_online_mask))
> > + cpumask_copy(pmask, possible_mask);
> > + }
> > +
> > + rcu_read_unlock();
> > spin_unlock_irqrestore(&callback_lock, flags);
>
> So, I suppose you're suggesting applying this patch instead of the one Will
> Deacon posted[1] and we need Will's second patch[2] on top, right?
>
> [1] http://lkml.kernel.org/r/20230131221719.3176-3-will@xxxxxxxxxx
> [2] http://lkml.kernel.org/r/20230131221719.3176-3-will@xxxxxxxxxx
FWIW, although I tend to share Peter's sentiments in this thread, I took
this (+ my second patch) for a spin and my tests are giving the same
results when compared with Peter's patch.
Tested-by: Will Deacon <will@xxxxxxxxxx>
Will
