[PATCH v5] ARM: new platform for Energy Micro's EFM32 Cortex-M3 SoCs

[Uwe Kleine-König]

Hello,

On Thu, Dec 12, 2013 at 01:55:38PM -0800, Olof Johansson wrote:
> On Thu, Dec 12, 2013 at 1:28 PM, Uwe Kleine-König
> <u.kleine-koenig at pengutronix.de> wrote:
> > On Wed, Dec 11, 2013 at 05:15:05PM -0800, Olof Johansson wrote:
> >> On Tue, Dec 10, 2013 at 05:55:16PM +0100, Uwe Kleine-König wrote:
> >> > There are still some missing parts (e.g. board support, device trees),
> >> > but with these bits added on top of this patch I can successfully boot a
> >> > EFM32GG-DK3750 board that uses an EFM32GG990F1024.
> >> >
> >> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig at pengutronix.de>
> >> > ---
> >> > Hello,
> >> >
> >> > I'd like to get this patch in for 3.14. The latest submission (v4,
> >> > Message-Id:1383562853-17202-4-git-send-email-u.kleine-koenig at pengutronix.de)
> >> > had some strict dependencies:
> >> >  - http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=7889/1
> >> >    to be able to drop Makefile.boot, this is still in Status=Incoming in
> >> >    rmk's patch tracker.
> >> >  - http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=7890/1
> >> >    to be able to drop the unused <mach/entry-macro.S>. This is already
> >> >    applied by rmk in his for-next branch, so it's not stable yet.
> >> >  - My series to drop <mach/timex.h> for !MULTI_PLATFORM. I'll send a
> >> >    pull request for that soon.
> >> >  - Patch "clocksource: time-efm32: Select CLKSRC_MMIO". This is part of
> >> >    a pull request by Daniel Lezcano that is expected to hit Linus T.'s
> >> >    tree for the next -rc. It currently sits as
> >> >    4ff3774e451fefb1781638c76230b3cf9a206b92 in Daniel's tree at
> >> >    git://git.linaro.org/people/daniel.lezcano/linux.git clockevents/fixes
> >> >    .
> >> >
> >> > To make application a bit easier I added dummy files to drop the
> >> > dependencies on the first three items of the list above.
> >> >
> >> > So it would be great if you could apply this patch when 4ff3774e4 went
> >> > in. (In case you want to apply it already now, this last dependency
> >> > could be work arounded, too, by adding a "select CLKSRC_MMIO" to
> >> > ARCH_EFM32.)
> >> >
> >> > Other than that the only change since v4 is that I mention the exact
> >> > board/processor combo in the changelog now.
> >> >
> >> > Thanks for considering this patch, best regards
> >>
> >> Applied to next/soc with a slightly tweaked patch subject.
> > Thanks, but I wonder if it's a problem that the commit doesn't build? (i.e.
> >
> >         git checkout f9b9cef1329c74607f33a20dce3f26e9a1f4ae96
> >         make efm32_defconfig
> >         make
> >
> > fails in
> >
> >  - arch/arm/kernel/sigreturn_codes.S with
> >    "Error: selected processor does not support ARM opcodes" and a few
> >    variations.
> >  - drivers/clocksource/time-efm32.o fails to link, because next/soc
> >    isn't a decendant of 4ff3774e451f.
> >
> > You can fix the latter by adding "select CLKSRC_MMIO" to ARCH_EFM32 if
> > you want. I don't remember what is the current state of the first
> > problem.)
> 
> Ok, dropped the patch again. Please let us know when the former has
> been resolved.
The former problem is fixed in 5091333 (ARM: 7895/1: signal: fix armv7-m
build issue in sigreturn_codes.S) which is in v3.13-rc3, but next/soc is
based on v3.13-rc1.

I suggest to wait until 4ff3774e451f hits Linus T.'s tree and then
reapply the patch.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |