From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752104AbdARJgh (ORCPT ); Wed, 18 Jan 2017 04:36:37 -0500 Received: from mx2.suse.de ([195.135.220.15]:34712 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751157AbdARJgf (ORCPT ); Wed, 18 Jan 2017 04:36:35 -0500 Date: Wed, 18 Jan 2017 10:34:48 +0100 From: Michal Hocko To: Vlastimil Babka Cc: Mel Gorman , Ganapatrao Kulkarni , linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [RFC 2/4] mm, page_alloc: fix fast-path race with cpuset update or removal Message-ID: <20170118093448.GI7015@dhcp22.suse.cz> References: <20170117221610.22505-1-vbabka@suse.cz> <20170117221610.22505-3-vbabka@suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170117221610.22505-3-vbabka@suse.cz> User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue 17-01-17 23:16:08, Vlastimil Babka wrote: > Ganapatrao Kulkarni reported that the LTP test cpuset01 in stress mode triggers > OOM killer in few seconds, despite lots of free memory. The test attemps to > repeatedly fault in memory in one process in a cpuset, while changing allowed > nodes of the cpuset between 0 and 1 in another process. > > One possible cause is that in the fast path we find the preferred zoneref > according to current mems_allowed, so that it points to the middle of the > zonelist, skipping e.g. zones of node 1 completely. If the mems_allowed is > updated to contain only node 1, we never reach it in the zonelist, and trigger > OOM before checking the cpuset_mems_cookie. > > This patch fixes the particular case by redoing the preferred zoneref search > if we switch back to the original nodemask. The condition is also slightly > changed so that when the last non-root cpuset is removed, we don't miss it. OK, the patch makes sense but longterm we really have to get rid of this insane switching between masks dances. > Note that this is not a full fix, and more patches will follow. > > Reported-by: Ganapatrao Kulkarni > Fixes: 682a3385e773 ("mm, page_alloc: inline the fast path of the zonelist iterator") > Signed-off-by: Vlastimil Babka Acked-by: Michal Hocko > --- > mm/page_alloc.c | 10 +++++++++- > 1 file changed, 9 insertions(+), 1 deletion(-) > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index 593a11d8bc6b..dedadb4a779f 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -3783,9 +3783,17 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, > /* > * Restore the original nodemask if it was potentially replaced with > * &cpuset_current_mems_allowed to optimize the fast-path attempt. > + * Also recalculate the starting point for the zonelist iterator or > + * we could end up iterating over non-eligible zones endlessly. > */ > - if (cpusets_enabled()) > + if (unlikely(ac.nodemask != nodemask)) { > ac.nodemask = nodemask; > + ac.preferred_zoneref = first_zones_zonelist(ac.zonelist, > + ac.high_zoneidx, ac.nodemask); > + if (!ac.preferred_zoneref->zone) > + goto no_zone; > + } > + > page = __alloc_pages_slowpath(alloc_mask, order, &ac); > > no_zone: > -- > 2.11.0 -- Michal Hocko SUSE Labs