[GIT PULL 3/3] ARM: shmobile: lager: enable Ether

[Olof Johansson]

On Thu, Aug 22, 2013 at 09:16:49AM +0900, Simon Horman wrote:
> On Wed, Aug 21, 2013 at 11:13:30PM +0400, Sergei Shtylyov wrote:
> > Hello.
> > 
> > On 08/21/2013 08:01 PM, Kevin Hilman wrote:
> > 
> > >>Acked-by: Kuninori Morimoto <kuninori.morimoto.gx at renesas.com>
> > >>Signed-off-by: Simon Horman <horms+renesas at verge.net.au>
> > 
> > >This one is triggering a new build failure in -next:
> > 
> > >arch/arm/mach-shmobile/board-lager.c:99:2: error: unknown field 'register_type' specified in initializer
> > >arch/arm/mach-shmobile/board-lager.c:99:19: error: 'SH_ETH_REG_FAST_RCAR' undeclared here (not in a function)
> > 
> > >>+/* Ether */
> > >>+static struct sh_eth_plat_data ether_pdata __initdata = {
> > >>+	.phy			= 0x1,
> > >>+	.edmac_endian		= EDMAC_LITTLE_ENDIAN,
> > >>+	.register_type		= SH_ETH_REG_FAST_RCAR,
> > 
> > >and it looks to be because this board is trying to use the register_type
> > >field which was removed by:
> > 
> > >commit 8d3214c4e8c8be6efd8ec7a172239ebbd4deb04b
> > >Author: Sergei Shtylyov <sergei.shtylyov at cogentembedded.com>
> > >Date:   Sun Aug 18 03:13:26 2013 +0400
> > >
> > >     sh_eth: remove 'register_type' field from 'struct sh_eth_plat_data'
> > >
> > >     Now that the 'register_type' field of the 'sh_eth' driver's platform data is not
> > >     used by the driver anymore, it's time to remove it and  its initializers from
> > >     the SH platform code. Also  move *enum* declaring values for this  field from
> > >     <linux/sh_eth.h>  to  the  local driver's  header file as they're only needed
> > >     by the driver itself  now...
> > >
> > >     Signed-off-by: Sergei Shtylyov <sergei.shtylyov at cogentembedded.com>
> > >     Signed-off-by: David S. Miller <davem at davemloft.net>
> > 
> >    Looks like in this case I had no choice since R8A7790 support for
> > the 'sh_eth' driver will be added in 3.12 only, so if I merged these
> > patches thru the shmobile tree, we'd have gotten an error in the
> > driver...
> >    I guess it's too early to fix this in any particular tree except
> > for linux-next itself... let me know what I can do.
> 
> This patch ("ARM: shmobile: lager: enable Ether") is also targeted at v3.12.
> So it seems to me that a fix for this could go through the shmobile tree
> if it is acceptable to merge net-next into the base of a pull-request.
> 
> Kevin, Olof, could you give some guidance here.

I don't think we want to pull in net-next here just for this -- that is a
massive branch.

There's no great solution to any of this, unfortunately.

I think easiest would be if we add a patch that fixes this up _only_ in
for-next, and merge the same patch upstream as soon as net-next lands
(if it lands after arm-soc, otherwise we include it in our branch),
that might be the least painful way to do this.

That patch would just remove the assignment to register_type in the
board file.


-Olof