Re: [GIT PULL 0/12] ARM: soc: updates for 3.7, first batch

[Olof Johansson]

Hi,

On Mon, Oct 01, 2012 at 07:22:34PM -0700, Linus Torvalds wrote:
> On Mon, Oct 1, 2012 at 4:56 PM, Olof Johansson <olof@xxxxxxxxx> wrote:
> >
> > Here is the first batch of pull requests for arm-soc for this merge
> > window. Some of them are tiny, and we will probably start to combine
> > them into slightly larger misc branches in the future. I didn't do it
> > this time around since the dependencies between branches ended up quite
> > complex this time around.
> 
> Ok, I've merged them all. As usual, my end result is slightly
> different from yours, although the differences are pretty small.

Thanks!

>  - I added exynos4210-trats.dtb to the "dtb-$(CONFIG_ARCH_EXYNOS)"
> case in arch/arm/boot/dts/Makefile
>
>    That seemed to be the right thing, otherwise that dtb file isn't
> actually accessed from anything. Hmm?

Hmm. I know I did that edit. I've had cases where git rerere has forgotten
about edits to "new" files that were not part of the original conflict, and
I might have replayed that merge and lost it there. Thanks for catching it.

>  - Commit dbc04161048d ("ARM: OMAP: Split plat/hardware.h, use local
> soc.h for omap2+") added the line
> 
>     #include "soc.h"
> 
>    to arch/arm/mach-omap2/omap_l3_noc.c, and I left it in as it moved
> into drivers/bus/.
> 
>    My resolution is quite possibly wrong, but I didn't know what the
> right thing to do was. Please double-check.

I had a patch for this but I forgot to move the tag to include it, my local
branch with the resolutions has it since I used the branch head, not the tag,
to merge it.

The SHA is a55eb3ec0e38e2db17207b204c050fbcdf750abb. Or the patch is included
below if that's easier.

>  - non-semantic order/whitespace difference for
> CONFIG_CLKSRC_ARM_GENERIC in drivers/clocksource/Makefile.
>
>    I didn't want to mix the generic case up with the random cases.

Yes, makes sense, thanks. I wonder if the file should be sorted to reduce
conflicts on new additions soon anyway, maybe.

> Anyway, you should double-check the merges, the "soc.h" one in
> particular is suspect.

Yes, my bad for not remembering to move the tag forward. The patch is below.

Thanks!

-Olof