From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756863Ab2DDPgI (ORCPT ); Wed, 4 Apr 2012 11:36:08 -0400 Received: from mail-qc0-f174.google.com ([209.85.216.174]:51990 "EHLO mail-qc0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756311Ab2DDPgF convert rfc822-to-8bit (ORCPT ); Wed, 4 Apr 2012 11:36:05 -0400 MIME-Version: 1.0 In-Reply-To: <20120404145134.GC12676@redhat.com> References: <20120403183655.GA23106@dhcp-172-17-108-109.mtv.corp.google.com> <20120404145134.GC12676@redhat.com> Date: Wed, 4 Apr 2012 10:36:04 -0500 Message-ID: Subject: Re: [Lsf] [RFC] writeback and cgroup From: Steve French To: Vivek Goyal Cc: Tejun Heo , ctalbott@google.com, rni@google.com, andrea@betterlinux.com, containers@lists.linux-foundation.org, linux-kernel@vger.kernel.org, lsf@lists.linux-foundation.org, linux-mm@kvack.org, jmoyer@redhat.com, lizefan@huawei.com, linux-fsdevel@vger.kernel.org, cgroups@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Apr 4, 2012 at 9:51 AM, Vivek Goyal wrote: > On Tue, Apr 03, 2012 at 11:36:55AM -0700, Tejun Heo wrote: > > Hi Tejun, > > Thanks for the RFC and looking into this issue. Few thoughts inline. > > [..] >> IIUC, without cgroup, the current writeback code works more or less >> like this.  Throwing in cgroup doesn't really change the fundamental >> design.  Instead of a single pipe going down, we just have multiple >> pipes to the same device, each of which should be treated separately. >> Of course, a spinning disk can't be divided that easily and their >> performance characteristics will be inter-dependent, but the place to >> solve that problem is where the problem is, the block layer. > > How do you take care of thorottling IO to NFS case in this model? Current > throttling logic is tied to block device and in case of NFS, there is no > block device. Similarly smb2 gets congestion info (number of "credits") returned from the server on every response - but not sure why congestion control is tied to the block device when this would create problems for network file systems -- Thanks, Steve From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steve French Subject: Re: [Lsf] [RFC] writeback and cgroup Date: Wed, 4 Apr 2012 10:36:04 -0500 Message-ID: References: <20120403183655.GA23106@dhcp-172-17-108-109.mtv.corp.google.com> <20120404145134.GC12676@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Tejun Heo , ctalbott-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org, rni-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org, andrea-oIIqvOZpAevzfdHfmsDf5w@public.gmane.org, containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, lsf-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, jmoyer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org, linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Vivek Goyal Return-path: In-Reply-To: <20120404145134.GC12676-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-fsdevel.vger.kernel.org On Wed, Apr 4, 2012 at 9:51 AM, Vivek Goyal wrote: > On Tue, Apr 03, 2012 at 11:36:55AM -0700, Tejun Heo wrote: > > Hi Tejun, > > Thanks for the RFC and looking into this issue. Few thoughts inline. > > [..] >> IIUC, without cgroup, the current writeback code works more or less >> like this. =A0Throwing in cgroup doesn't really change the fundament= al >> design. =A0Instead of a single pipe going down, we just have multipl= e >> pipes to the same device, each of which should be treated separately= =2E >> Of course, a spinning disk can't be divided that easily and their >> performance characteristics will be inter-dependent, but the place t= o >> solve that problem is where the problem is, the block layer. > > How do you take care of thorottling IO to NFS case in this model? Cur= rent > throttling logic is tied to block device and in case of NFS, there is= no > block device. Similarly smb2 gets congestion info (number of "credits") returned from the server on every response - but not sure why congestion control is tied to the block device when this would create problems for network file systems --=20 Thanks, Steve From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx206.postini.com [74.125.245.206]) by kanga.kvack.org (Postfix) with SMTP id 801EB6B007E for ; Wed, 4 Apr 2012 11:36:05 -0400 (EDT) Received: by qadb15 with SMTP id b15so549231qad.9 for ; Wed, 04 Apr 2012 08:36:04 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20120404145134.GC12676@redhat.com> References: <20120403183655.GA23106@dhcp-172-17-108-109.mtv.corp.google.com> <20120404145134.GC12676@redhat.com> Date: Wed, 4 Apr 2012 10:36:04 -0500 Message-ID: Subject: Re: [Lsf] [RFC] writeback and cgroup From: Steve French Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-linux-mm@kvack.org List-ID: To: Vivek Goyal Cc: Tejun Heo , ctalbott@google.com, rni@google.com, andrea@betterlinux.com, containers@lists.linux-foundation.org, linux-kernel@vger.kernel.org, lsf@lists.linux-foundation.org, linux-mm@kvack.org, jmoyer@redhat.com, lizefan@huawei.com, linux-fsdevel@vger.kernel.org, cgroups@vger.kernel.org On Wed, Apr 4, 2012 at 9:51 AM, Vivek Goyal wrote: > On Tue, Apr 03, 2012 at 11:36:55AM -0700, Tejun Heo wrote: > > Hi Tejun, > > Thanks for the RFC and looking into this issue. Few thoughts inline. > > [..] >> IIUC, without cgroup, the current writeback code works more or less >> like this. =A0Throwing in cgroup doesn't really change the fundamental >> design. =A0Instead of a single pipe going down, we just have multiple >> pipes to the same device, each of which should be treated separately. >> Of course, a spinning disk can't be divided that easily and their >> performance characteristics will be inter-dependent, but the place to >> solve that problem is where the problem is, the block layer. > > How do you take care of thorottling IO to NFS case in this model? Current > throttling logic is tied to block device and in case of NFS, there is no > block device. Similarly smb2 gets congestion info (number of "credits") returned from the server on every response - but not sure why congestion control is tied to the block device when this would create problems for network file systems --=20 Thanks, Steve -- 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/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: email@kvack.org