On Thu, Oct 04, 2018 at 10:10:48AM +0100, Quentin Perret wrote:
> On Wednesday 03 Oct 2018 at 18:27:19 (+0200), Peter Zijlstra wrote:
> > On Wed, Sep 12, 2018 at 10:13:03AM +0100, Quentin Perret wrote:
> > > @@ -288,6 +321,21 @@ static void build_perf_domains(const struct cpumask
> > > *cpu_map)
> > > goto free;
> > > tmp->next = pd;
> > > pd = tmp;
> > > +
> > > + /*
> > > + * Count performance domains and capacity states for the
> > > + * complexity check.
> > > + */
> > > + nr_pd++;
> > > + nr_cs += em_pd_nr_cap_states(pd->obj);
> > > + }
> > > +
> > > + /* Bail out if the Energy Model complexity is too high. */
> > > + if (nr_pd * (nr_cs + nr_cpus) > EM_MAX_COMPLEXITY) {
> > > + if (sched_debug())
> > > + pr_info("rd %*pbl: EM complexity is too high\n ",
> > > + cpumask_pr_args(cpu_map));
> > > + goto free;
> > > }
> >
> > I would make than an unconditional WARN, we do not really expect that to
> > trigger, but then it does, we really don't want to hide it.
>
> OTOH that also means that some people with big asymmetric machines can
> get a WARN message every time they boot, and even if they don't want to
> use EAS.
>
> Now, that shouldn't happen any time soon, so it's maybe a good thing if
> we get reports when/if people start to hit that one, so why not ...
Right, and if becomes a real problem we can think of a solution (like
maybe a DT thingy that says to not use EAS, or a 'better' EAS
algorithm).
