From mboxrd@z Thu Jan 1 00:00:00 1970 From: Glauber Costa Subject: Re: [PATCH v2 10/11] memcg: allow a memcg with kmem charges to be destructed. Date: Wed, 22 Aug 2012 12:36:17 +0400 Message-ID: <50349A01.5020906@parallels.com> References: <1344517279-30646-1-git-send-email-glommer@parallels.com> <1344517279-30646-11-git-send-email-glommer@parallels.com> <20120821082259.GB19797@dhcp22.suse.cz> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20120821082259.GB19797@dhcp22.suse.cz> Sender: owner-linux-mm@kvack.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Michal Hocko Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, cgroups@vger.kernel.org, devel@openvz.org, Johannes Weiner , Andrew Morton , kamezawa.hiroyu@jp.fujitsu.com, Christoph Lameter , David Rientjes , Pekka Enberg , Pekka Enberg , Suleiman Souhlal On 08/21/2012 12:22 PM, Michal Hocko wrote: > On Thu 09-08-12 17:01:18, Glauber Costa wrote: >> Because the ultimate goal of the kmem tracking in memcg is to track slab >> pages as well, we can't guarantee that we'll always be able to point a >> page to a particular process, and migrate the charges along with it - >> since in the common case, a page will contain data belonging to multiple >> processes. >> >> Because of that, when we destroy a memcg, we only make sure the >> destruction will succeed by discounting the kmem charges from the user >> charges when we try to empty the cgroup. > > This changes the semantic of memory.force_empty file because the usage > should be 0 on success but it will show kmem usage in fact now. I guess > it is inevitable with u+k accounting so you should be explicit about > that and also update the documentation. aaand, it's done. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org