From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?Q?Bj=C3=B8rn_Mork?= Subject: Re: Using interface number probing for usbnet drivers before 3.6-rc1? Date: Wed, 04 Jul 2012 16:50:53 +0200 Message-ID: <87ipe3lgvm.fsf@nemi.mork.no> References: <1340282758-18085-1-git-send-email-bjorn@mork.no> <20120622.171735.637204872051168180.davem@davemloft.net> <87y5neidk1.fsf@nemi.mork.no> <20120623.023749.1066357959466872561.davem@davemloft.net> <87mx3flqns.fsf_-_@nemi.mork.no> <20120704144649.GC23020@kroah.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , netdev@vger.kernel.org, linux-usb@vger.kernel.org To: Greg Kroah-Hartman Return-path: Received: from canardo.mork.no ([148.122.252.1]:40190 "EHLO canardo.mork.no" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752593Ab2GDOvD convert rfc822-to-8bit (ORCPT ); Wed, 4 Jul 2012 10:51:03 -0400 In-Reply-To: <20120704144649.GC23020@kroah.com> (Greg Kroah-Hartman's message of "Wed, 4 Jul 2012 07:46:49 -0700") Sender: netdev-owner@vger.kernel.org List-ID: Greg Kroah-Hartman writes: > On Wed, Jul 04, 2012 at 01:19:35PM +0200, Bj=C3=B8rn Mork wrote: >> David Miller writes: >> > From: Bj=C3=B8rn Mork >> > Date: Sat, 23 Jun 2012 11:26:54 +0200 >> > >> >> Is waiting for 3.6-rc1 the best alternative, or is cherry-picking= them >> >> an option? >> > >> > I'm extremely reluctant to have a dependency on another tree. >> > >> > The USB tree, on the other hand, could pull the net-next tree >> > in so you can continue your dependent work there. >>=20 >> Hello Greg, >>=20 >> I know I said I would be more careful about cross-tree dependencies,= but >> here I am again... >>=20 >> The context is: How should I proceed with interface number based pro= bing >> for usbnet drivers, in particular qmi_wwan which already has a few >> patches in net-next? >>=20 >> I see two possible remaining options: >>=20 >> 1) you pull the net-next changes into usb-next so that I can do the >> work against usb-next now, or >> 2) I wait for 3.6-rc1 and do the work against the net tree after >> 3.6-rc1 is merged there >>=20 >> I have decided to go for option #2 unless either of you have other >> preferences, as I believe that is going to cause the least work and >> merge trouble for you both. > > What about: > 3) You send the usb interface number matching patches again for th= em > to be included in the net-next tree as well, causing duplicates to= be > in linux-next, but allowing you to send your patches into net-next= =2E > > That is what we have done at times for other patches, and if David is= ok > with it, so am I. Odds are, the merge will happen automatically, but= if > not, I can handle it when 3.6-rc1 comes around. > > Sound good? I could have misunderstood, but I thought that was out of the question given that David didn't want to cherry-pick them from usb-next. Bj=C3=B8rn=20