[PATCH 3/6] ARM: remove fa526 CPU support

[kaloz@openwrt.org (Imre Kaloz)]

On Tue, 19 Mar 2013 15:55:18 +0100, Imre Kaloz <kaloz@openwrt.org> wrote:

> On Tue, 19 Mar 2013 11:35:45 +0100, Arnd Bergmann <arnd@arndb.de> wrote:
>
>> On Tuesday 19 March 2013, Imre Kaloz wrote:
>>> I have two wbd111 boards, but their bootloader is fubared - for whatever
>>> reason the bootloader's ethernet support works horribly for me, as in an
>>> image transfer takes well over 10 mins, and 90% of the time it fails
>>> meantime. I have another gemini based board, which doesn't have mainline
>>> support, yet.
>>>
>>> So for gemini, what I can offer is digging up the wbd111 and testing your
>>> patches this weekend - if that works, it can stay in mainline for another
>>> kernel release, and I can try to add support for the other board (so
>>> there would be a sane platform to test on). If you are going to drop it,
>>> I will likely keep it working in OpenWrt as long as those boards don't
>>> die on me.
>>
>> Sounds good. If you are still interested in keeping the code alive, we
>> shouldn't drop it upstream. Can you also add a defconfig file that
>> enables all the gemini specific Kconfig code plus whatever you use
>> in OpenWRT?
>>
>> Part of the reason why the platform was so broken is that it never had
>> a defconfig, and we didn't notice the bitrot.
>
> Sure, I'll try to get all of this done this weekend.
>

Just tested 3.9-rc4 with your "ARM: gemini: get platform to build again" patch and the ones we have in our tree and it works on my WBD111. Please apply that patch and with a little luck I can get the other board working for 3.11, with the additional defconfig and some more hw support as well -- or if you would like so, I can get a defconfig done for 3.10, but I'm not sure it makes too much sense.


Imre