Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753469Ab3C0PcZ (ORCPT ); Wed, 27 Mar 2013 11:32:25 -0400 Received: from cantor2.suse.de ([195.135.220.15]:55852 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752611Ab3C0PcW (ORCPT ); Wed, 27 Mar 2013 11:32:22 -0400 Date: Wed, 27 Mar 2013 16:32:20 +0100 From: Michal Hocko To: Glauber Costa Cc: Johannes Weiner , Tejun Heo , Li Zefan , KAMEZAWA Hiroyuki , linux-mm@kvack.org, cgroups@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] memcg: fix memcg_cache_name() to use cgroup_name() Message-ID: <20130327153220.GL16579@dhcp22.suse.cz> References: <1364373399-17397-1-git-send-email-mhocko@suse.cz> <20130327145727.GD29052@cmpxchg.org> <20130327151104.GK16579@dhcp22.suse.cz> <51530E1E.3010100@parallels.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <51530E1E.3010100@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: 1652 Lines: 44 On Wed 27-03-13 19:19:58, Glauber Costa wrote: > On 03/27/2013 07:11 PM, Michal Hocko wrote: > > On Wed 27-03-13 10:58:25, Johannes Weiner wrote: > >> On Wed, Mar 27, 2013 at 09:36:39AM +0100, Michal Hocko wrote: > > [...] > >>> + /* > >>> + * kmem_cache_create_memcg duplicates the given name and > >>> + * cgroup_name for this name requires RCU context. > >>> + * This static temporary buffer is used to prevent from > >>> + * pointless shortliving allocation. > >>> + */ > >>> + if (!tmp_name) { > >>> + tmp_name = kmalloc(PAGE_SIZE, GFP_KERNEL); > >>> + WARN_ON_ONCE(!tmp_name); > >> > >> Just use the page allocator directly and get a free allocation failure > >> warning. > > > > WARN_ON_ONCE is probably pointless. > > > >> Then again, order-0 pages are considered cheap enough that they never > >> even fail in our current implementation. > >> > >> Which brings me to my other point: why not just a simple single-page > >> allocation? > > > > No objection from me. I was previously thinking about the "proper" > > size for something that is a file name. So I originally wanted to use > > PATH_MAX instead but ended up with PAGE_SIZE for reasons I do not > > remember now. > > theoretically, this is PATH_MAX + max cache name. So do you prefer kmalloc(PATH_MAX) or the page allocator directly as Johannes suggests? I agree tha kamlloc(PAGE_SIZE) looks weird. -- 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/