From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752732AbcEMNe4 (ORCPT ); Fri, 13 May 2016 09:34:56 -0400 Received: from smtpoutz26.laposte.net ([194.117.213.101]:34051 "EHLO smtp.laposte.net" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751104AbcEMNez (ORCPT ); Fri, 13 May 2016 09:34:55 -0400 Message-ID: <5735D7FC.3070409@laposte.net> Date: Fri, 13 May 2016 15:34:52 +0200 From: Sebastian Frias User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.2.0 MIME-Version: 1.0 To: "Austin S. Hemmelgarn" , Michal Hocko CC: Mason , linux-mm@kvack.org, Andrew Morton , Linus Torvalds , LKML Subject: Re: [PATCH] mm: add config option to select the initial overcommit mode References: <5731CC6E.3080807@laposte.net> <20160513080458.GF20141@dhcp22.suse.cz> <573593EE.6010502@free.fr> <5735A3DE.9030100@laposte.net> <20160513120042.GK20141@dhcp22.suse.cz> <5735CAE5.5010104@laposte.net> <935da2a3-1fda-bc71-48a5-bb212db305de@gmail.com> In-Reply-To: <935da2a3-1fda-bc71-48a5-bb212db305de@gmail.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-VR-SrcIP: 83.142.147.193 X-VR-FullState: 0 X-VR-Score: -100 X-VR-Cause-1: gggruggvucftvghtrhhoucdtuddrfeekledrvddugdeigecutefuodetggdotefrodftvfcurfhrohhf X-VR-Cause-2: ihhlvgemucfntefrqffuvffgnecuuegrihhlohhuthemucehtddtnecusecvtfgvtghiphhivghnthhs X-VR-Cause-3: ucdlqddutddtmdenucfjughrpefkfffhfgggvffufhgjtgfgsehtjegrtddtfeehnecuhfhrohhmpefu X-VR-Cause-4: vggsrghsthhirghnucfhrhhirghsuceoshhfkeegsehlrghpohhsthgvrdhnvghtqeenucfkphepkeef X-VR-Cause-5: rddugedvrddugeejrdduleefnecurfgrrhgrmhepmhhouggvpehsmhhtphhouhhtpdhhvghloheplgdu X-VR-Cause-6: jedvrddvjedrtddrvddugegnpdhinhgvthepkeefrddugedvrddugeejrdduleefpdhmrghilhhfrhho X-VR-Cause-7: mhepshhfkeegsehlrghpohhsthgvrdhnvghtpdhrtghpthhtoheprghhfhgvrhhrohhinhejsehgmhgr X-VR-Cause-8: ihhlrdgtohhm X-VR-AvState: No X-VR-State: 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Austin, On 05/13/2016 03:11 PM, Austin S. Hemmelgarn wrote: > On 2016-05-13 08:39, Sebastian Frias wrote: >> >> My point is that it seems to be possible to deal with such conditions in a more controlled way, ie: a way that is less random and less abrupt. > There's an option for the OOM-killer to just kill the allocating task instead of using the scoring heuristic. This is about as deterministic as things can get though. By the way, why does it has to "kill" anything in that case? I mean, shouldn't it just tell the allocating task that there's not enough memory by letting malloc return NULL? Best regards, Sebastian