Re: linux-next: manual merge of the mfd tree with the v4l-dvb tree

[Mauro Carvalho Chehab <mchehab@redhat.com>]

Em Tue, 16 Apr 2013 21:23:59 +0200
Samuel Ortiz <sameo@linux.intel.com> escreveu:

> On Tue, Apr 16, 2013 at 09:25:45AM -0700, Andrey Smirnov wrote:
> > On Tue, Apr 16, 2013 at 6:51 AM, Samuel Ortiz <sameo@linux.intel.com> wrote:
> > > Hi Mauro,
> > >
> > > On Wed, Apr 10, 2013 at 06:48:28AM -0300, Mauro Carvalho Chehab wrote:
> > >> Em Wed, 10 Apr 2013 08:42:53 +0200
> > >> Samuel Ortiz <sameo@linux.intel.com> escreveu:
> > >>
> > >> > Hi Stephen,
> > >> >
> > >> > On Wed, Apr 10, 2013 at 01:48:13PM +1000, Stephen Rothwell wrote:
> > >> > > Hi Samuel,
> > >> > >
> > >> > > Today's linux-next merge of the mfd tree got a conflict in
> > >> > > drivers/mfd/Kconfig between commit 3f8ec5df11aa ("[media] mfd: Add header
> > >> > > files and Kbuild plumbing for SI476x MFD core") from the v4l-dvb tree and
> > >> > > commit ab85b120e692 ("mfd: Kconfig alphabetical re-ordering") from the
> > >> > > mfd tree.
> > >> > I'm surprised the v4l-dvb tree is carrying this patchset because I haven't
> > >> > ACKed it.
> > >>
> > >> Sorry. Not sure why I understood that you gave your ack. Perhaps I miss-read
> > >> one of the comments of that thread.
> > > I haven't heard back from Andrey yet. If I don't get anything from him on
> > > Wednesday, would you mind reverting this patchset ? I have not ACKed it
> > > because it breaks bisectability, and it conflicts with mfd-next.
> > >
> > > Andrey, any plans to adress my comments from last week ?
> > 
> > I have a new version of a patchset that addresses your comments and
> > incorporates a couple of other patches with bugfixes. Unfortunately I
> > haven't had a chance to run it on the test HW I have. I'll try to do
> > and post the patches this week, but I am not sure if I'll have time to
> > do it before Wednesday.
> Linus will most likely tag 3.9 on Sunday or so, I won't take any patches after
> that (And probably no patches after Friday...)
> 
>  
> > As far as I understand all the MFD patches will go through the "mfd"
> > tree which doesn't have any version of the SI476X related patches and
> > I don't have to worry about making incremental patches. This, however,
> > is not the case for "media_tree" this patch here
> > http://git.linuxtv.org/media_tree.git/commit/30bac9110455402fa8888740c6819dd3daa2666f
> > would be affected and I think that making the need changes
> > incrementally as a separate patch would break the bisectability also.
> > Right now what I have is a new version of said patch produced by
> > editing the history(rebase and squash). Would that work for you,
> > Mauro, or do you want an incremental patch on top of the original one?
> If you manage to send the patches on time, here is what I propose: I create a
> stable branch with all your mfd patches and merge it on my master branch.
> Mauro can then also merge it and then apply the rest of your patchset (which
> is all v4l and media related, iirc). Before doing so, Mauro would have to
> revert the last patchset from you, that's currently sitting in his tree.

Ok, reverted both changesets 3f8ec5df11aa and 30bac9110455402f:

http://git.linuxtv.org/media_tree.git/commit/82cd0b278fddc1c0bc7e187ff82fd0e273520233
http://git.linuxtv.org/media_tree.git/commit/33a31edd4a4b7d26b962b32decfd8ea2377eaa0d

By reverting first 30bac9110455402f, I avoided breaking git bisect.

Regards,
Mauro