[PATCH v2 03/11] ARM: EXYNOS: add clock part for EXYNOS5250 SoC

Turquette, Mike mturquette at ti.com
Thu Feb 16 13:38:36 EST 2012


On Wed, Feb 15, 2012 at 9:50 PM, Arnd Bergmann <arnd at arndb.de> wrote:
> On Thursday 16 February 2012, Turquette, Mike wrote:
>> >
>> >> Ah, one more, would be better to us if arm-soc tree could provide
>> >> the topic branch for 'common struct clk' working as a base.
>> >
>> > Good point. Mike, can you send a pull request for whatever you have
>> > now as another staging branch for arm-soc?
>>
>> Arnd, we discussed linux-next as the target for the common clk core
>> code at Linaro Connect.  Are you now asking for that code in arm-soc?
>> If so, I think it would be better to keep things simple only target
>> linux-next once V5 has hit the lists after ELC.
>
> I guess waiting for you to send out v5 is a good idea, but as Kgene
> mentioned it would be nice to have the series in arm-soc in order
> to base other branches on top of it, and we have now made it possible
> to update branches like this one by declaring them "staging".
>
> The main advantage that I see of putting your series into arm-soc
> is that I can handle the dependencies: If you want to update the
> series based to v6 after more review and send me a replacement pull
> request, I can rebase the exynos5 patches (and other conversions)
> on top of that. If you have a different tree in linux-next and plan
> to rebase it, I cannot take any other patches into arm-soc that depend
> on yours.

Per our conversation at ELC, we can host in arm-soc to make it easier
for folks to port.  But the code should likely be merged through
Russell once we have the final version.

Regards,
Mike

> The alternative would be that I take the exynos5 patches in their
> current shape into arm-soc and you put your patches into linux-next
> based on the branch that I have, and with a patch to convert exynos5
> to it on top.
>
>> If you are instead referring to OMAP platform support for common clk,
>> that code is very infantile and not yet ready for arm-soc, especially
>> as it breaks OMAP2+ multi-boot.
>
> I did not mean the omap specific parts.
>
>        Arnd



More information about the linux-arm-kernel mailing list