From mboxrd@z Thu Jan 1 00:00:00 1970 From: Oliver Neukum Subject: Re: [PATCH] cdc-ether: clean packet filter upon probe Date: Mon, 25 Aug 2014 14:13:42 +0200 Message-ID: <1408968822.10300.18.camel@linux-fkkt.site> References: <1406537796-20156-1-git-send-email-oneukum@suse.de> <87a977z5pn.fsf@nemi.mork.no> <1408083121.17565.4.camel@linux-fkkt.site> <87d2c2xm2k.fsf@nemi.mork.no> <1408954063.10300.1.camel@linux-fkkt.site> <87a96tc692.fsf@nemi.mork.no> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: davem@davemloft.net, netdev@vger.kernel.org To: =?ISO-8859-1?Q?Bj=F8rn?= Mork Return-path: Received: from cantor2.suse.de ([195.135.220.15]:46425 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751731AbaHYMNo (ORCPT ); Mon, 25 Aug 2014 08:13:44 -0400 In-Reply-To: <87a96tc692.fsf@nemi.mork.no> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, 2014-08-25 at 10:39 +0200, Bj=C3=B8rn Mork wrote: > Oliver Neukum writes: > > On Fri, 2014-08-15 at 09:13 +0200, Bj=C3=B8rn Mork wrote: > >> Oliver Neukum writes: > >>=20 > >> > On Thu, 2014-08-14 at 13:11 +0200, Bj=C3=B8rn Mork wrote: > > > >> > Support for this request is mandatory. > >>=20 > >> Yes, there is no problem for standard conforming ECM devices. I f= ully > >> agree there.=20 > >>=20 > >> But standard conformance is unfortunately not something we can exp= ect. > >> My question was regarding the non-conforming devices, like the one= s > >> supported by th zaurus and rndis_host drivers. These drivers are = made > > > > You have a point. Yet another flag? Do you see an alternative? >=20 > Note that I don't know whether this is a real problem or not. I don't > believe I have any zaurus or rndis_host devices. >=20 > But how about just splitting the .bind functions for true ECM devices > and the others, letting the zaurus and rndis_host driver continue to = use > usbnet_cdc_bind with no filter setting? Maybe define a new unexporte= d > .bind in cdc_ether for this purpose, moving the ECM specific > initialization there? >=20 > Maybe something along this (instead of the change to > usbnet_generic_cdc_bind)?: Yes, that looks superior. Regards Oliver