Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933681Ab2JLIog (ORCPT ); Fri, 12 Oct 2012 04:44:36 -0400 Received: from cantor2.suse.de ([195.135.220.15]:35001 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753005Ab2JLIoe (ORCPT ); Fri, 12 Oct 2012 04:44:34 -0400 Date: Fri, 12 Oct 2012 10:44:32 +0200 From: Michal Hocko To: Glauber Costa Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, Andrew Morton , Mel Gorman , Suleiman Souhlal , Tejun Heo , cgroups@vger.kernel.org, kamezawa.hiroyu@jp.fujitsu.com, Johannes Weiner , Greg Thelen , devel@openvz.org, Frederic Weisbecker Subject: Re: [PATCH v4 14/14] Add documentation about the kmem controller Message-ID: <20121012084431.GF10110@dhcp22.suse.cz> References: <1349690780-15988-1-git-send-email-glommer@parallels.com> <1349690780-15988-15-git-send-email-glommer@parallels.com> <20121011143559.GJ29295@dhcp22.suse.cz> <5077CC73.80504@parallels.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5077CC73.80504@parallels.com> 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 Content-Length: 1477 Lines: 38 On Fri 12-10-12 11:53:23, Glauber Costa wrote: > On 10/11/2012 06:35 PM, Michal Hocko wrote: > > On Mon 08-10-12 14:06:20, Glauber Costa wrote: [...] > >> Kernel memory limits are not imposed for the root cgroup. Usage for the root > >> -cgroup may or may not be accounted. > >> +cgroup may or may not be accounted. The memory used is accumulated into > >> +memory.kmem.usage_in_bytes, or in a separate counter when it makes sense. > > > > Which separate counter? Is this about tcp kmem? > > > > So far, yes, this is the only case that makes sense, and the fewer the > better. In any case it exists, and I wanted to be generic. Add (currently tcp) or something similar [...] > >> + Kernel memory is effectively set as a percentage of the user memory. This > > > > not a percentage it is subset of the user memory > > > Well, this is semantics. I can change, but for me it makes a lot of > sense to think of it in terms of a percentage, because it is easy to > administer. You don't actually write a percentage, which I tried to > clarify by using the term "effective set as a percentage". I can still see somebody reading this and wondering why echo 50 > ...limit didn't set a percentage... -- Michal Hocko SUSE Labs -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/