From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754583Ab1CWHr4 (ORCPT ); Wed, 23 Mar 2011 03:47:56 -0400 Received: from fgwmail5.fujitsu.co.jp ([192.51.44.35]:55386 "EHLO fgwmail5.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752981Ab1CWHrz (ORCPT ); Wed, 23 Mar 2011 03:47:55 -0400 X-SecurityPolicyCheck-FJ: OK by FujitsuOutboundMailChecker v1.3.1 Date: Wed, 23 Mar 2011 16:41:22 +0900 From: KAMEZAWA Hiroyuki To: KOSAKI Motohiro Cc: linux-kernel@vger.kernel.org, Andrew Morton , David Rientjes , Linus Torvalds , Rik van Riel , Oleg Nesterov , linux-mm , Andrey Vagin , Hugh Dickins , Nick Piggin , Minchan Kim , Johannes Weiner Subject: Re: [PATCH 1/5] vmscan: remove all_unreclaimable check from direct reclaim path completely Message-Id: <20110323164122.ea25bdf0.kamezawa.hiroyu@jp.fujitsu.com> In-Reply-To: <20110322200523.B061.A69D9226@jp.fujitsu.com> References: <20110315153801.3526.A69D9226@jp.fujitsu.com> <20110322194721.B05E.A69D9226@jp.fujitsu.com> <20110322200523.B061.A69D9226@jp.fujitsu.com> Organization: FUJITSU Co. LTD. X-Mailer: Sylpheed 3.1.0 (GTK+ 2.10.14; i686-pc-mingw32) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 22 Mar 2011 20:05:55 +0900 (JST) KOSAKI Motohiro wrote: > all_unreclaimable check in direct reclaim has been introduced at 2.6.19 > by following commit. > > 2006 Sep 25; commit 408d8544; oom: use unreclaimable info > > And it went through strange history. firstly, following commit broke > the logic unintentionally. > > 2008 Apr 29; commit a41f24ea; page allocator: smarter retry of > costly-order allocations > > Two years later, I've found obvious meaningless code fragment and > restored original intention by following commit. > > 2010 Jun 04; commit bb21c7ce; vmscan: fix do_try_to_free_pages() > return value when priority==0 > > But, the logic didn't works when 32bit highmem system goes hibernation > and Minchan slightly changed the algorithm and fixed it . > > 2010 Sep 22: commit d1908362: vmscan: check all_unreclaimable > in direct reclaim path > > But, recently, Andrey Vagin found the new corner case. Look, > > struct zone { > .. > int all_unreclaimable; > .. > unsigned long pages_scanned; > .. > } > > zone->all_unreclaimable and zone->pages_scanned are neigher atomic > variables nor protected by lock. Therefore a zone can become a state > of zone->page_scanned=0 and zone->all_unreclaimable=1. In this case, > current all_unreclaimable() return false even though > zone->all_unreclaimabe=1. > > Is this ignorable minor issue? No. Unfortunatelly, x86 has very > small dma zone and it become zone->all_unreclamble=1 easily. and > if it becase all_unreclaimable, it never return all_unreclaimable=0 > beucase it typicall don't have reclaimable pages. > > Eventually, oom-killer never works on such systems. Let's remove > this problematic logic completely. > > Reported-by: Andrey Vagin > Cc: Nick Piggin > Cc: Minchan Kim > Cc: Johannes Weiner > Cc: Rik van Riel > Cc: KAMEZAWA Hiroyuki > Signed-off-by: KOSAKI Motohiro IIUC, I saw the pehnomenon which you pointed out, as - all zone->all_unreclaimable = yes - zone_reclaimable() returns true - no pgscan proceeds. on a swapless system. So, I'd like to vote for this patch. But hmm...what happens all of pages are isolated or locked and now under freeing ? I think we should have alternative safe-guard logic for avoiding to call oom-killer. Hmm. Thanks, -Kame