From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cong Wang Subject: Re: [PATCH] tun: teach the tun/tap driver to support netpoll Date: Thu, 16 Jun 2011 09:44:39 +0800 Message-ID: <4DF96007.50904@redhat.com> References: <1308151501-8434-1-git-send-email-nhorman@tuxdriver.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, Rik van Riel , Maxim Krasnyansky , "David S. Miller" To: Neil Horman Return-path: Received: from mx1.redhat.com ([209.132.183.28]:37164 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752385Ab1FPBpG (ORCPT ); Wed, 15 Jun 2011 21:45:06 -0400 In-Reply-To: <1308151501-8434-1-git-send-email-nhorman@tuxdriver.com> Sender: netdev-owner@vger.kernel.org List-ID: =E4=BA=8E 2011=E5=B9=B406=E6=9C=8815=E6=97=A5 23:25, Neil Horman =E5=86= =99=E9=81=93: > Commit 8d8fc29d02a33e4bd5f4fa47823c1fd386346093 changed the behavior = of slave > devices in regards to netpoll. Specifically it created a mutually ex= clusive > relationship between being a slave and a netpoll-capable device. Thi= s creates > problems for KVM because guests relied on needing netconsole active o= n a slave > device to a bridge. Ideally libvirtd could just attach netconsole to= the bridge > device instead, but thats currently infeasible, because while the bri= dge device > supports netpoll, it requires that all slave interface also support i= t, but the > tun/tap driver currently does not. The most direct solution is to te= ach tun/tap > to support netpoll, which is implemented by the patch below. > Yeah, bridge checks if all the underlying devices support netpoll, that is why Rik saw the failure. > I've not tested this yet, but its pretty straightforward. > > Signed-off-by: Neil Horman > Reported-by: Rik van Riel > CC: Rik van Riel > CC: Maxim Krasnyansky > CC: Cong Wang > CC: "David S. Miller" Reviewed-by: WANG Cong Thanks for fixing it!