Re: [PATCH 5/7] mm, vmscan: extract shrink_page_list reclaim counters into a struct

[Michal Hocko]

On Wed 04-01-17 15:51:43, Vlastimil Babka wrote:
[...]
> > @@ -1266,11 +1270,13 @@ static unsigned long shrink_page_list(struct list_head *page_list,
> >  	list_splice(&ret_pages, page_list);
> >  	count_vm_events(PGACTIVATE, pgactivate);
> >  
> > -	*ret_nr_dirty += nr_dirty;
> > -	*ret_nr_congested += nr_congested;
> > -	*ret_nr_unqueued_dirty += nr_unqueued_dirty;
> > -	*ret_nr_writeback += nr_writeback;
> > -	*ret_nr_immediate += nr_immediate;
> > +	if (stat) {
> > +		stat->nr_dirty = nr_dirty;
> > +		stat->nr_congested = nr_congested;
> > +		stat->nr_unqueued_dirty = nr_unqueued_dirty;
> > +		stat->nr_writeback = nr_writeback;
> > +		stat->nr_immediate = nr_immediate;
> > +	}
> 
> This change of '+=' to '=' raised my eybrows, but it seems both callers
> don't care so this is indeed no functional change and potentially faster.

Yes, I was quite surprised as well, maybe we had a code which relied on
the aggregated numbers in the past but I didn't bother to go over git
logs to check. There is no such user anymore...
 

-- 
Michal Hocko
SUSE Labs