[U-Boot] [PATCH v3 1/4] kconfig: arm: introduce symbol for ARM CPUs

[Masahiro YAMADA]

Hi Albert,



2014-10-28 16:01 GMT+09:00 Albert ARIBAUD <albert.u.boot at aribaud.net>:
> Hello Masahiro,
>
> On Tue, 28 Oct 2014 02:23:42 +0900, Masahiro YAMADA
> <yamada.m at jp.panasonic.com> wrote:
>> Hi Albert,
>>
>>
>> 2014-10-28 2:09 GMT+09:00 Albert ARIBAUD <albert.u.boot at aribaud.net>:
>> >> Tom and Albert,
>> >>
>> >> If Georges fixes those problems, is it possible to apply it shortly on
>> >> the mainline?
>> >> (directly u-boot/master? or via u-boot-arm/master?)
>> >
>> > I'm ok with this patch going in (through either tree). However:
>> >
>> >> Since this patch easily causes conflicts, it should not get stuck for
>> >> a long time.
>> >
>> > I'm not sure I'm getting this right. Do yo mean that once the patch is
>> > in, it should be easy to detect and iron out issues because if some
>> > boards are broken, their users will react promptly?
>>
>> Yes.
>>
>> If some conflicts happen, CONFIG_SYS_CPU will not be set correctly
>> and those boards will fail to build.
>> User will notice the issue soon.
>
> I often run 'buildman ... aarch64 arm' as a measure of the cleanliness
> of the ARM repo. Would just applying this patch automatically cause lots
> of ARM boards to fail building?


If some conflicts happen after merging this patch and something goes wrong,
I think you can _probably_ detect it by buildman, but I am not 100% sure.


 How about moving forward like this?

  1. Georges will update his patch.

  2. I will test it asap.
     (guarantee the correctness by comparing MD5SUM for all the ARM boards.)

  3. Tom will immediately apply it to u-boot/master.

  4. Announce every custodian should get his repo synced with the upstream.

  5. We may still have some conflicts, so we will need to fix them carefully.
     (This fix won't be so difficult.  We just have to focus on some
new boards.)


In this way, I think we have only a couple of conflicts even if we have.

-- 
Best Regards
Masahiro Yamada