[GIT PULL v2] efm32 cleanups for next

Olof Johansson olof at lixom.net
Sun Feb 23 08:51:51 EST 2014


On Sat, Feb 22, 2014 at 10:58 AM, Uwe Kleine-König
<u.kleine-koenig at pengutronix.de> wrote:
> Hello Olof,
>
> On Fri, Feb 21, 2014 at 04:38:53PM -0800, Olof Johansson wrote:
>> On Thu, Feb 20, 2014 at 08:54:52PM +0100, Uwe Kleine-König wrote:
>> > Hello,
>> >
>> > this pull request superseeds the earlier request this mail is a reply
>> > to. Compared to the original one I added another cleanup. Here come the
>> > full details:
>> >
>> > The following changes since commit 11d73c56b96b4297bd96273e9a22b7c72d13f1fa:
>> >
>> >   Merge tag 'dropmachtimexh-v2' of git://git.pengutronix.de/git/ukl/linux into next/cleanup (2014-02-18 22:19:33 -0800)
>> >
>> > are available in the git repository at:
>> >
>> >   git://git.pengutronix.de/git/ukl/linux.git tags/efm32-next
>> >
>> > for you to fetch changes up to f41b0482d5c1fc05234d54d58931a606ec7a97a2:
>> >
>> >   ARM: efm32: drop selecting CLKSRC_MMIO (2014-02-20 09:56:18 +0100)
>> >
>> > ----------------------------------------------------------------
>> > Just three simple cleanups for efm32 removing two now unused files and a
>> > superflous select by ARCH_EFM32.
>> >
>> > ----------------------------------------------------------------
>> > BTW, I based this on armsoc/cleanup because this is the first commit
>> > that allows to remove efm32's <mach/timex.h> (which is the last one in
>> > the tree).
>>
>> We ask you to never base your branches on top of our next/* contents, since it
>> happens that it will be rebuilt. Instead you should base your branch on top of
>> the topic branch in question, i.e. your timex-removal branch.
> my timex-removal branch isn't enough, because it bases on v3.13-rc1
> which does neither have arch/arm/mach-efm32 nor the commits that allow
> to drop the select CLKSRC_MMIO nor the commit that allows to drop
> <mach/entry-macro.S>.
>
> That's why I wrote that armsoc/[next/]cleanup is the first commit that
> makes my cleanups possible.
>
> The only options I see are:
>  1) merge v3.14-rc1 into my timex-series myself and base on that;
>  2) wait until my timex-series enters a tree with >= 3.14-rc1 that is
>     stable; or
>  3) drop commit "ARM: efm32: drop unused file <mach/timex.h>" for now
>     and base the rest on top of 3.14-rc1.
>
> I did 2) assuming that next/cleanup is stable. Actually this isn't much
> different from 1) because the commit I based my changes on is just
> 3.14-rc1 + your merge of my series. So even if you want to redo
> next/cleanup throwing out some merges different than mine it shouldn't
> be too hard to keep the merge you did.
>
> The details are:
>
>  - "ARM: efm32: drop selecting CLKSRC_MMIO" depends on
>         09ca27579ee5 (clocksource: time-efm32: Select CLKSRC_MMIO)
>    which is in v3.13-rc5;
>  - "ARM: efm32: drop unused file <mach/timex.h>" depends on
>         79f08d9ed217 (ARM: drop <mach/timex.h> for !ARCH_MULTIPLATFORM, too)
>         6d85e2b0b6be (ARM: new platform for Energy Micro's EFM32 Cortex-M3 SoCs)
>    so I need v3.14-rc1 + dropmachtimexh-v2 for it;
>  - "ARM: efm32: drop unused file entry-macro.S" depends on
>         e6d159cdf9f7 (ARM: 7890/1: v7-M: drop using mach/entry-macro.S)
>    which is in v3.14-rc1.
>
> Please advice,

Ok, it's probably easiest to just apply the patches directly for us,
so I'll do that.

If your branch is in linux-next through your own tree then you should
drop it once the patches hit our tree. I'll likely do this later today
at some point.


-Olof



More information about the linux-arm-kernel mailing list