From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ding Tianhong Subject: Re: [PATCH net-next 0/2] macvlan: fix the problem when mac address changes for passthru mode Date: Thu, 29 May 2014 11:29:58 +0800 Message-ID: <5386A9B6.40505@huawei.com> References: <1401106233-11320-1-git-send-email-dingtianhong@huawei.com> <53860DA5.8090407@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: To: Vlad Yasevich , , Return-path: Received: from szxga02-in.huawei.com ([119.145.14.65]:63657 "EHLO szxga02-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753458AbaE2DbW (ORCPT ); Wed, 28 May 2014 23:31:22 -0400 In-Reply-To: <53860DA5.8090407@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On 2014/5/29 0:24, Vlad Yasevich wrote: > On 05/26/2014 08:10 AM, Ding Tianhong wrote: >> Ding Tianhong (2): >> macvlan: don't change macvlan address alone for passthru mode >> macvlan: propagate lowerdev mac address changes for passthru mode >> >> drivers/net/macvlan.c | 40 +++++++++++++++++++++++++++++++--------- >> 1 file changed, 31 insertions(+), 9 deletions(-) >> > > This ends up working kind-of backwards... A user now has to know > that passthru macvlans he has to change the lower-level mac, but > for all others, changing the macvlan mac is ok and will work... Yes, more rules for it, but I think a user should know it. > > This also will not work if the passthru macvlan is stacked on > something other the hw (for example: a vlan). passthru is a special mode, will not be used for every situation. > > Why not simply propagate the mac address change of the passhtru macvlan > down to the real device? > Yes, I am considering this problem for a long time, I'm not sure the macvlan should propagate the mac address change to the real device, but your suggestion looks good to me, thanks for your advise, I will think about it and fix it. Ding > Thanks > -vlad > >