From mboxrd@z Thu Jan 1 00:00:00 1970 From: "tj@kernel.org" Subject: Re: [Lsf-pc] [LSF/MM ATTEND] Filesystems -- Btrfs, cgroups, Storage topics from Facebook Date: Thu, 2 Jan 2014 12:10:09 -0500 Message-ID: <20140102171009.GJ11501@htj.dyndns.org> References: <1388439412.16965.27.camel@ret> <20131231084927.GA29449@gmail.com> <20131231124535.GE11920@quack.suse.cz> <1388495991.16965.36.camel@ret> <52C2D342.8000606@tao.ma> <1388504116.24668.0.camel@ret.masoncoding.com> <20140102064659.GF11920@quack.suse.cz> <1388676106.24668.14.camel@ret.masoncoding.com> <20140102160102.GH11501@htj.dyndns.org> <20140102170637.GA13276@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Chris Mason , "jack@suse.cz" , "lizefan@huawei.com" , "gnehzuil.liu@gmail.com" , "tm@tao.ma" , "lsf-pc@lists.linux-foundation.org" , "linux-fsdevel@vger.kernel.org" To: Vivek Goyal Return-path: Received: from mail-qe0-f41.google.com ([209.85.128.41]:37260 "EHLO mail-qe0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753777AbaABRKN (ORCPT ); Thu, 2 Jan 2014 12:10:13 -0500 Received: by mail-qe0-f41.google.com with SMTP id gh4so14667508qeb.28 for ; Thu, 02 Jan 2014 09:10:12 -0800 (PST) Content-Disposition: inline In-Reply-To: <20140102170637.GA13276@redhat.com> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Hey, Vivek. On Thu, Jan 02, 2014 at 12:06:37PM -0500, Vivek Goyal wrote: > So is this a separate configuration which can be done per bdi as opposed > to per device? IOW throttling offered per per cgroup per bdi. This will > help with the case of throttling over NFS too, which some people have > been asking for. Hah? No, bdi just being split per-cgroup on each device so that it can properly propagate congestion upwards per-blkcg, just like how we split request allocation per-cgroup in the block layer proper. > So it sounds like re-implementing throttling infrastructure at bdi level > now (Similar to what has been done at device level)? Of course use as > much code as possible. But IIUC, proposal is that effectively there will > can be two throttling controllers. One operating at bdi level and one > operating below it at device level? Not at all. I was arguing explicitly against that. Thanks. -- tejun