Re: [PATCH 3/4] mm: unreserve highatomic free pages fully before OOM

[Michal Hocko <mhocko@kernel.org>]

On Tue 11-10-16 16:37:16, Minchan Kim wrote:
> On Tue, Oct 11, 2016 at 09:26:06AM +0200, Michal Hocko wrote:
> > On Tue 11-10-16 16:09:45, Minchan Kim wrote:
[...]
> > > @@ -2154,12 +2156,24 @@ static void unreserve_highatomic_pageblock(const struct alloc_context *ac)
> > >  			 * may increase.
> > >  			 */
> > >  			set_pageblock_migratetype(page, ac->migratetype);
> > > -			move_freepages_block(zone, page, ac->migratetype);
> > > -			spin_unlock_irqrestore(&zone->lock, flags);
> > > -			return;
> > > +			ret = move_freepages_block(zone, page,
> > > +						ac->migratetype);
> > > +			/*
> > > +			 * By race with page freeing functions, !highatomic
> > > +			 * pageblocks can have free pages in highatomic free
> > > +			 * list so if drain is true, try to unreserve every
> > > +			 * free pages in highatomic free list without bailing
> > > +			 * out.
> > > +			 */
> > > +			if (!drain) {
> > 
> > 			if (ret)
> > > +				spin_unlock_irqrestore(&zone->lock, flags);
> > > +				return ret;
> > > +			}
> > 
> > arguably this would work better also for !drain case which currently
> > tries to unreserve but in case of the race it would do nothing.
> 
> I thought it but I was afraid if you say again it's over complicated.

Well, maybe there is even better/easier solution. Anyway, if
I were you I would just split it into two patches. The first
to unreserve from shoudl_reclaim_retry and the later to make
unreserve_highatomic_pageblock more reliable.

> I will do it with your SOB in next spin.

ok, thanks!
-- 
Michal Hocko
SUSE Labs