From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753201AbbANO2w (ORCPT ); Wed, 14 Jan 2015 09:28:52 -0500 Received: from mx2.parallels.com ([199.115.105.18]:38082 "EHLO mx2.parallels.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751272AbbANO2v (ORCPT ); Wed, 14 Jan 2015 09:28:51 -0500 Date: Wed, 14 Jan 2015 17:28:41 +0300 From: Vladimir Davydov To: Johannes Weiner CC: Andrew Morton , Michal Hocko , Greg Thelen , , , Subject: Re: [patch 2/2] mm: memcontrol: default hierarchy interface for memory Message-ID: <20150114142841.GE11264@esperanza> References: <1420776904-8559-1-git-send-email-hannes@cmpxchg.org> <1420776904-8559-2-git-send-email-hannes@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Disposition: inline In-Reply-To: <1420776904-8559-2-git-send-email-hannes@cmpxchg.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jan 08, 2015 at 11:15:04PM -0500, Johannes Weiner wrote: > - memory.low configures the lower end of the cgroup's expected > memory consumption range. The kernel considers memory below that > boundary to be a reserve - the minimum that the workload needs in > order to make forward progress - and generally avoids reclaiming > it, unless there is an imminent risk of entering an OOM situation. AFAICS, if a cgroup cannot be shrunk back to its low limit (e.g. because it consumes anon memory, and there's no swap), it will get on with it. Is it considered to be a problem? Are there any plans to fix it, e.g. by invoking OOM-killer in a cgroup that is above its low limit if we fail to reclaim from it? Thanks, Vladimir