Re: [PATCH 03/27] mm, vmstat: Add infrastructure for per-node vmstats

[Mel Gorman]

On Tue, Feb 23, 2016 at 10:13:18AM -0800, Johannes Weiner wrote:
> On Tue, Feb 23, 2016 at 03:04:26PM +0000, Mel Gorman wrote:
> > VM statistic counters for reclaim decisions are zone-based. If the kernel
> > is to reclaim on a per-node basis then we need to track per-node statistics
> > but there is no infrastructure for that. The most notable change is that
> > the old node_page_state is renamed to sum_zone_node_page_state.  The new
> > node_page_state takes a pglist_data and uses per-node stats but none exist
> > yet. There is some renaming such as vm_stat to vm_zone_stat and the addition
> > of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise,
> > this is mostly a mechanical patch with no functional change. There is a
> > lot of similarity between the node and zone helpers which is unfortunate
> > but there was no obvious way of reusing the code and maintaining type safety.
> > 
> > Signed-off-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
> 
> Hopefully we can eventually ditch /proc/zoneinfo in favor of a
> /proc/nodeinfo and get rid of the per-zone stats accounting.
> 

It may not be possible to ditch /proc/zoneinfo entirely but a /proc/nodeinfo
would make sense. It may interfere with userspace that's aware of kernel
internals but that may be manageable.

> In general, this patch looks good to me.
> 
> Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
> 

Thanks.

> Only one thing I noticed:
> 
> > @@ -349,12 +349,14 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker,
> >  	shadow_nodes = list_lru_shrink_count(&workingset_shadow_nodes, sc);
> >  	local_irq_enable();
> >  
> > -	if (memcg_kmem_enabled())
> > +	if (memcg_kmem_enabled()) {
> >  		pages = mem_cgroup_node_nr_lru_pages(sc->memcg, sc->nid,
> >  						     LRU_ALL_FILE);
> > -	else
> > -		pages = node_page_state(sc->nid, NR_ACTIVE_FILE) +
> > -			node_page_state(sc->nid, NR_INACTIVE_FILE);
> > +	} else {
> > +		pg_data_t *pgdat = NODE_DATA(sc->nid);
> > +		pages = node_page_state(pgdat, NR_ACTIVE_FILE) +
> > +			node_page_state(pgdat, NR_INACTIVE_FILE);
> > +	}
> 
> That should also be sum_zone_node_page_state, right? These are not
> valid node items (yet).

Yep, not for another two patches. Fixed now.

-- 
Mel Gorman
SUSE Labs