[PATCH 0/2] Fix legacy boards compilation breakage with DU

Arnd Bergmann arnd at arndb.de
Mon Dec 1 07:09:23 PST 2014


On Friday 28 November 2014 10:17:33 Laurent Pinchart wrote:
> On Friday 28 November 2014 09:18:46 Simon Horman wrote:
> > [CCed Magnus, ARM SoC maintainers]
> > 
> > On Thu, Nov 27, 2014 at 05:19:11PM +0200, Laurent Pinchart wrote:
> > > Hello,
> > > 
> > > The DU driver has lost support for platform data, resulting in a
> > > compilation breakage for the legacy Marzen and Lager board files that
> > > managed to keep under the radar until now.
> > > 
> > > As the multiplatform boards should be used instead, drop support for DU in
> > > the legacy Marzen and Lager boards.
> > > 
> > > Simon, this is required to fix a compilation breakage in the drm-next
> > > branch. I'm sorry for not catching it earlier :-/ How would you like this
> > > to go in ? Could the patches be applied to the DRM tree ? They seem to
> > > apply cleanly to both drm-next and your latest devel branch.
> > > 
> > > Laurent Pinchart (2):
> > >   ARM: shmobile: lager: Remove DU platform device
> > >   ARM: shmobile: marzen: Remove DU platform device
> > >  
> > >  arch/arm/mach-shmobile/board-lager.c  | 58 ------------------------------
> > >  arch/arm/mach-shmobile/board-marzen.c | 58 ------------------------------
> >
> > Yes, I think that should be fine as these fines are in maintenance mode and
> > the only recent change I see to them is to modify the FSF address, which
> > seems far away from the hunks in this patch-set.
> > 
> > In other words, I think the chance of conflicts is small and I am fine with
> > these changes going through the DRM tree if that is where the breakage
> > manifests.
> > 
> > Acked-by: Simon Horman <horms+renesas at verge.net.au>
> > 
> > I am of course, happy to take them if the DRM maintainer(s) prefer me to.
> 
> Dave, how would you prefer to handle this ?

If you want to pick it up, you can also add my 

Acked-by: Arnd Bergmann <arnd at arndb.de>

Otherwise we can merge it through arm-soc, but that would cause a
larger bisection problem.


More information about the dri-devel mailing list