From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755304AbbGCRO1 (ORCPT ); Fri, 3 Jul 2015 13:14:27 -0400 Received: from mail-yk0-f176.google.com ([209.85.160.176]:36110 "EHLO mail-yk0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755143AbbGCROX (ORCPT ); Fri, 3 Jul 2015 13:14:23 -0400 Date: Fri, 3 Jul 2015 13:14:19 -0400 From: Tejun Heo To: Jan Kara Cc: axboe@kernel.dk, linux-kernel@vger.kernel.org, hch@infradead.org, hannes@cmpxchg.org, linux-fsdevel@vger.kernel.org, vgoyal@redhat.com, lizefan@huawei.com, cgroups@vger.kernel.org, linux-mm@kvack.org, mhocko@suse.cz, clm@fb.com, fengguang.wu@intel.com, david@fromorbit.com, gthelen@google.com, khlebnikov@yandex-team.ru Subject: Re: [PATCH 22/51] writeback: add {CONFIG|BDI_CAP|FS}_CGROUP_WRITEBACK Message-ID: <20150703171419.GG5273@mtj.duckdns.org> References: <1432329245-5844-1-git-send-email-tj@kernel.org> <1432329245-5844-23-git-send-email-tj@kernel.org> <20150630093751.GH7252@quack.suse.cz> <20150702011056.GC26440@mtj.duckdns.org> <20150703104957.GH23329@quack.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150703104957.GH23329@quack.suse.cz> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello, On Fri, Jul 03, 2015 at 12:49:57PM +0200, Jan Kara wrote: > Well, unless there is some specific mapping for the device, we could just > fall back to attributing everything to the root cgroup. We would still > account dirty pages in memcg, throttle writers in memcg when there are too > many dirty pages, issue writeback for inodes in memcg with enough dirty > pages etc. Just all IO from different memcgs would be equal so no > separation would be there. But it would still seem better that just > ignoring the split of dirty pages among memcgs as we do now... Thoughts? Sure, if you mark a bdi as capable of supporing cgroup writeback without enforcing any IO isolation, the above would be what's happening. I'm not convinced this would be something actually useful tho. Sure, it changes the behavior but is still gonna be a crapshoot. Thanks. -- tejun