Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932403AbcCHMWw (ORCPT ); Tue, 8 Mar 2016 07:22:52 -0500 Received: from mail-wm0-f48.google.com ([74.125.82.48]:35755 "EHLO mail-wm0-f48.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752861AbcCHMWo (ORCPT ); Tue, 8 Mar 2016 07:22:44 -0500 Date: Tue, 8 Mar 2016 13:22:41 +0100 From: Michal Hocko To: Vlastimil Babka Cc: Hugh Dickins , Sergey Senozhatsky , Andrew Morton , Linus Torvalds , Johannes Weiner , Mel Gorman , David Rientjes , Tetsuo Handa , Hillf Danton , KAMEZAWA Hiroyuki , linux-mm@kvack.org, LKML , Joonsoo Kim Subject: Re: [PATCH] mm, oom: protect !costly allocations some more Message-ID: <20160308122241.GD13542@dhcp22.suse.cz> References: <20160203132718.GI6757@dhcp22.suse.cz> <20160225092315.GD17573@dhcp22.suse.cz> <20160229210213.GX16930@dhcp22.suse.cz> <20160307160838.GB5028@dhcp22.suse.cz> <56DE9A68.2010301@suse.cz> <20160308094612.GB13542@dhcp22.suse.cz> <56DEA0CF.2070902@suse.cz> <20160308101016.GC13542@dhcp22.suse.cz> <56DEB394.40602@suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <56DEB394.40602@suse.cz> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2907 Lines: 66 On Tue 08-03-16 12:12:20, Vlastimil Babka wrote: > On 03/08/2016 11:10 AM, Michal Hocko wrote: > > On Tue 08-03-16 10:52:15, Vlastimil Babka wrote: > >> On 03/08/2016 10:46 AM, Michal Hocko wrote: > > [...] > >>>>> @@ -3294,6 +3289,18 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, > >>>>> did_some_progress > 0, no_progress_loops)) > >>>>> goto retry; > >>>>> > >>>>> + /* > >>>>> + * !costly allocations are really important and we have to make sure > >>>>> + * the compaction wasn't deferred or didn't bail out early due to locks > >>>>> + * contention before we go OOM. > >>>>> + */ > >>>>> + if (order && order <= PAGE_ALLOC_COSTLY_ORDER) { > >>>>> + if (compact_result <= COMPACT_CONTINUE) > >>>> > >>>> Same here. > >>>> I was going to say that this didn't have effect on Sergey's test, but > >>>> turns out it did :) > >>> > >>> This should work as expected because compact_result is unsigned long > >>> and so this is the unsigned arithmetic. I can make > >>> #define COMPACT_NONE -1UL > >>> > >>> to make the intention more obvious if you prefer, though. > >> > >> Well, what wasn't obvious to me is actually that here (unlike in the > >> test above) it was actually intended that COMPACT_NONE doesn't result in > >> a retry. But it makes sense, otherwise we would retry endlessly if > >> reclaim couldn't form a higher-order page, right. > > > > Yeah, that was the whole point. An alternative would be moving the test > > into should_compact_retry(order, compact_result, contended_compaction) > > which would be CONFIG_COMPACTION specific so we can get rid of the > > COMPACT_NONE altogether. Something like the following. We would lose the > > always initialized compact_result but this would matter only for > > order==0 and we check for that. Even gcc doesn't complain. > > Yeah I like this version better, you can add my Acked-By. OK, patch updated and I will post it as a reply to the original email. > Thanks. > > > A more important question is whether the criteria I have chosen are > > reasonable and reasonably independent on the particular implementation > > of the compaction. I still cannot convince myself about the convergence > > here. Is it possible that the compaction would keep returning > > compact_result <= COMPACT_CONTINUE while not making any progress at all? > > Theoretically, if reclaim/compaction suitability decisions and > allocation attempts didn't match the watermark checks, including the > alloc_flags and classzone_idx parameters. Possible scenarios: > > - reclaim thinks compaction has enough to proceed, but compaction thinks > otherwise and returns COMPACT_SKIPPED > - compaction thinks it succeeded and returns COMPACT_PARTIAL, but > allocation attempt fails > - and perhaps some other combinations But that might happen right now as well so it wouldn't be a regression, right? -- Michal Hocko SUSE Labs