Re: [PATCH 1/4] numa: introduce per-cgroup numa balancing locality, statistic

[Peter Zijlstra]

On Fri, Jul 12, 2019 at 11:43:17AM +0800, çè wrote:
> 
> 
> On 2019/7/11 äå9:47, Peter Zijlstra wrote:
> [snip]
> >> +	rcu_read_lock();
> >> +	memcg = mem_cgroup_from_task(p);
> >> +	if (idx != -1)
> >> +		this_cpu_inc(memcg->stat_numa->locality[idx]);
> > 
> > I thought cgroups were supposed to be hierarchical. That is, if we have:
> > 
> >           R
> > 	 / \
> > 	 A
> > 	/\
> > 	  B
> > 	  \
> > 	   t1
> > 
> > Then our task t1 should be accounted to B (as you do), but also to A and
> > R.
> 
> I get the point but not quite sure about this...
> 
> Not like pages there are no hierarchical limitation on locality, also tasks

You can use cpusets to affect that.

> running in a particular group have no influence to others, not to mention the
> extra overhead, does it really meaningful to account the stuff hierarchically?

AFAIU it's a requirement of cgroups to be hierarchical. All our other
cgroup accounting is like that.