From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760020AbcALQEc (ORCPT ); Tue, 12 Jan 2016 11:04:32 -0500 Received: from www62.your-server.de ([213.133.104.62]:40695 "EHLO www62.your-server.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752882AbcALQEa (ORCPT ); Tue, 12 Jan 2016 11:04:30 -0500 Message-ID: <5695240A.1020803@iogearbox.net> Date: Tue, 12 Jan 2016 17:04:26 +0100 From: Daniel Borkmann User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.4.0 MIME-Version: 1.0 To: Lubomir Rintel CC: netdev@vger.kernel.org, linux-kernel@vger.kernel.org, "David S. Miller" , Jamal Hadi Salim Subject: Re: [PATCH 1/3] ifb: make device count build-time configurable References: <1452599804-27284-1-git-send-email-lkundrak@v3.sk> <5695199E.2040303@iogearbox.net> <56951D49.5020800@iogearbox.net> <1452613429.18559.69.camel@v3.sk> In-Reply-To: <1452613429.18559.69.camel@v3.sk> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit X-Authenticated-Sender: daniel@iogearbox.net Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/12/2016 04:43 PM, Lubomir Rintel wrote: > On Tue, 2016-01-12 at 16:35 +0100, Daniel Borkmann wrote: >> On 01/12/2016 04:19 PM, Daniel Borkmann wrote: >>> On 01/12/2016 12:56 PM, Lubomir Rintel wrote: > ... >>>> -static int numifbs = 2; >>> ^^^^^^^^^^^^ >>>> +static int numifbs = CONFIG_IFB_COUNT; >>>> module_param(numifbs, int, 0); >>>> MODULE_PARM_DESC(numifbs, "Number of ifb devices"); >> >> Btw, perhaps there should also be a pr_info() or the like to warn the user >> that relying on modprobe creating these devs is deprecated and scheduled >> for removal in future, so they should switch to ip link instead ... > > I don't know, is it actually is scheduled for removal? > > I was under impression that changes in default behavior users can rely > on are generally near impossible to do since the potential problems > largely outweigh the benefits. The distributions will do the changes in > their kernels anyway and changing behavior on kernel upgrades in random > old systems might be risky. BLK_DEV_LOOP_MIN_COUNT is 8 by default on upstream today, 0 in Fedora kernels since Fedora 19 (2013) onwards [1] (F18 to F19 made the switch apparently). If the same happens one day to these options here and all major distros play along, then perhaps after still long enough time it might be an option (and if not, people at least switched to use ip link ;)). > That said, it's just an impression. I don't follow the development > closely enough to be sure about anything. [1] http://pkgs.fedoraproject.org/cgit/rpms/kernel.git/tree/config-generic?h=f19