[fedora-arm] Who's using Kirkwood?

[Peter Robinson]

On Sat, Oct 6, 2012 at 10:43 AM, Jon Masters <jcm at redhat.com> wrote:
> Hi Folks,
>
> I'm interested to know who is using Kirkwood, and who would miss it if
> it went away. For now, we won't kill off ARMv5 because it is used in the
> official rPi builds but that doesn't mean I'm not interested to know
> whether we should put testing effort into Kirkwood for F18.
>
> My thought is that the latest plugs are moving to ARMv7, and so as the
> cutting edge Linux distro, we should make plans for deprecating support
> over the coming releases. This is not a call to drop support today. If I
> can get numbers on how many people care, that will help.

Jon you have such a terrible way with words!

To explain what I believe Jon is trying to say a little better let me
outline my thoughts.

ARMv5 as a chip is going away, ARM is actively moving customers (their
customers not end users) to other ARMv7 based chips such as the
Cortex-A5 and A7 that have better performance and use let power for
similar costs. I don't believe there will be new products on ARMv5 by
the new year.

There's a lot of interesting new boards appearing on the market at a
less than $100 price point such as the A10 based Cubieboard [1] at $49
that has a 1gb of RAM and 1ghz Cortex-A8 chipset with things like real
SATA. There's also devices like the Wandboard [2] which gets you a
dual core A9 processor with 1gb of RAM for $89

So my thoughts have been what to do with armv5tel support, we're
certainly not going to tear it out. With Seneca moving to do a bringup
for the armv6hl architecture for the likes of the Pi there will be a
lot less users of armv5tel and the actual users of it other than the
Pi are low.

So my thoughts are that when we move to primary architecture that only
ARMv7 gets promoted. I believe this is best for the movement of ARM in
Fedora in general. Even just the promotion to primary cause some
heated discussion and I strongly suspect the release of Fedora that
will end up being the release that we push to primary will be Fedora
20. F-19 is already well underway and I believe we need to enable a
new arch at the point where the previous release branches which means
F-20 is out next opportunity. That means we need to look at what will
be on the market in 12 months time and I don't think ARMv5 will be on
the shopping list.

We merge the armv5tel koji infrastructure with that of the v6hl and it
remains as a secondary arch and continues as it is if those people
that are interested in it step up to maintain it. I will personally
continue to assist to ensure packages don't fail etc. Seneca have
provided a solid and stable infrastructure and they use the
architectures as a core component of their various courses and it
provides their students a fabulous learning environment for getting
into different platforms to x86 and to learn how to build and OS from
the ground up!

Of course none of this is set in stone, it's a discussion and just me
putting my ideas into words.

Peter

[1] http://cubieboard.org
[2] http://www.wandboard.org/