Re: [PATCH REPOST RFC cgroup/for-3.7] cgroup: mark subsystems with broken hierarchy support and whine if cgroups are nested for them

[Michal Hocko <mhocko@suse.cz>]

On Tue 11-09-12 10:07:46, Tejun Heo wrote:
> Hello, Michal.
> 
> On Tue, Sep 11, 2012 at 12:04:33PM +0200, Michal Hocko wrote:
> > >  	cgroup_unlock();
> > > @@ -4953,6 +4958,7 @@ mem_cgroup_create(struct cgroup *cont)
> > >  						&per_cpu(memcg_stock, cpu);
> > >  			INIT_WORK(&stock->work, drain_local_stock);
> > >  		}
> > > +		mem_cgroup_subsys.broken_hierarchy = !memcg->use_hierarchy;
> > 
> > Hmmm, this will warn even if we have
> > root (default use_hierarchy=0)
> >  \
> >   A (use_hierarchy=1)
> >    \
> >     B <- here
> > 
> > which is unfortunate because it will add a noise to a reasonable
> > configuration.
> 
> I suppose you're talking about having root group not performing any
> accounting and/or control?

It doesn't do any controlling because you cannot set any limit for it.
Root cgroup has always been special.

> I suppose such could be a valid use case
> (is it really necessary tho?)  but I don't think .use_hierarchy is the
> right interface for that.  

I am not sure I understand what you mean by that. My only concern with
is that we shouldn't complain if somebody doesn't do anything wrong.
And creating a group under root without any other children, no matter
what use_hierarchy says, is a valid use case and we shouldn't make too
much noise about that.
The only difference in such a use case is that hierarchical stats will
include numbers from the group only if root had use_hierarchy==1. There
are no other side effects.

> If it's absolutely necessary, I think it should be a root-only flag
> (even if that ends up using the same code path).  Eventually, we
> really want to kill .use_hierarchy, or at least make it to RO 1.  As
> it's currently defined, it's just way too confusing.

Agreed on that, definitely.

-- 
Michal Hocko
SUSE Labs