From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: Bisect'ed BUG in VLAN promisc mode (6c78dcbd47) Date: Fri, 26 Sep 2008 21:28:35 +0200 Message-ID: <48DD37E3.9090706@trash.net> References: <1222437636.7598.14.camel@localhost.localdomain> <48DD0A4F.6020703@trash.net> <1222456933.2381.3.camel@localhost.localdomain> <48DD36D2.90103@trash.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: "netdev@vger.kernel.org" To: jdb@comx.dk Return-path: Received: from stinky.trash.net ([213.144.137.162]:46187 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752857AbYIZT2h (ORCPT ); Fri, 26 Sep 2008 15:28:37 -0400 In-Reply-To: <48DD36D2.90103@trash.net> Sender: netdev-owner@vger.kernel.org List-ID: Patrick McHardy wrote: > Jesper Dangaard Brouer wrote: >> On Fri, 2008-09-26 at 18:14 +0200, Patrick McHardy wrote: >>> Jesper Dangaard Brouer wrote: >>>> Hi Patrick, >>>> >>>> Bisected me down to one of you changes >>>> commit 6c78dcbd47a68a >>>> [VLAN]: Fix promiscous/allmulti synchronization races >>>> >>>> Description: >>>> ------------ >>>> All other VLAN interfaces stop working, if a vlan is taken down >>>> (ifconfig eth1.1025 down) _while_ there is a tcpdump running on that >>>> interface. >>>> >>>> The problem is a result of promisc mode being removed on the >>>> real-device (eth1), when the vlan interface is taken down. This >>>> should not happen as other vlan devices exists that still need >>>> promisc mode on the real-device (eth1). >>> I'm pretty sure we fixed this already in commit 0ed21b32. >> >> I think that I have tested including this commit. The first thing I did >> was to test with latest DaveM net-2.6 git tree. > > Let me see if I can reproduce it ... Actually - one question: you're saying you're using different MAC addresses on the VLAN devices, so I guess thats why you're expecting the underlying device to still be in promiscous mode after you set eth1.1025 down. For devices that support multiple unicast addresses in hardware, we don't put the device in promiscous mode anymore. So the question is: is something actually not working, or did you just notice that the real device is no longer in promiscous mode?