From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030834Ab2K3PhT (ORCPT ); Fri, 30 Nov 2012 10:37:19 -0500 Received: from cantor2.suse.de ([195.135.220.15]:47384 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1030753Ab2K3PhR (ORCPT ); Fri, 30 Nov 2012 10:37:17 -0500 Date: Fri, 30 Nov 2012 16:37:15 +0100 From: Michal Hocko To: azurIt Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, cgroups mailinglist , KAMEZAWA Hiroyuki , Johannes Weiner Subject: Re: [PATCH for 3.2.34] memcg: do not trigger OOM from add_to_page_cache_locked Message-ID: <20121130153715.GK29317@dhcp22.suse.cz> References: <20121125120524.GB10623@dhcp22.suse.cz> <20121125135542.GE10623@dhcp22.suse.cz> <20121126013855.AF118F5E@pobox.sk> <20121126131837.GC17860@dhcp22.suse.cz> <20121126132149.GD17860@dhcp22.suse.cz> <20121130032918.59B3F780@pobox.sk> <20121130124506.GH29317@dhcp22.suse.cz> <20121130144427.51A09169@pobox.sk> <20121130144431.GI29317@dhcp22.suse.cz> <20121130150347.GJ29317@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20121130150347.GJ29317@dhcp22.suse.cz> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri 30-11-12 16:03:47, Michal Hocko wrote: [...] > Anyway, the more interesting thing is gfp_mask is GFP_NOWAIT allocation > from the page fault? Huh this shouldn't happen - ever. OK, it starts making sense now. The message came from pagefault_out_of_memory which doesn't have gfp nor the required node information any longer. This suggests that VM_FAULT_OOM has been returned by the fault handler. So this hasn't been triggered by the page fault allocator. I am wondering whether this could be caused by the patch but the effect of that one should be limitted to the write (unlike the later version for -mm tree which hooks into the shmem as well). Will have to think about it some more. -- Michal Hocko SUSE Labs