Karsten Hopp píše v Po 17. 05. 2010 v 18:03 +0200:
Am 17.05.2010 17:29, schrieb Dennis Gilmore:
> while it is not a complete tree i think we should move the rawhide tree to the
> fedora 12 location. it would be great if we could get some hercules images
> based on it. We can then work to try get F-13 and F-14 started
> I want to make sure that for all arches we release F-14 in the same month as
> the primary arch is released.
> There is plenty for people to do to get things moving forward. adding s390 31
> bit builds has slowed things down.
> I also wonder if we shouldn't make a 31 bit installer and kernel for fedora.
> or is there really no demand for it?
Phil created the last hercules images, F-11 afaik. Maybe he's got some
time to update them to F-12.
I will try to update the Phil's image to F-12 during the next few days.
If I remember there were some obstacles, but my own image is F-12 based,
so I should be able to resolve them again.
The big question atm is if we really should try to get an installable
There's still a lot of work to do, a rough estimate is 1 - 1/2 months and
then we'd play catch-up again with F-14. Phil and I talked about it in
the office today and we agreed that it might be a good idea to start
rebuilding Rawhide as soon as possible even if this means that we'd have to
rebuild a package twice to fix SONAME problems.
I agree that we should jump directly to rawhide. Otherwise we would be
always only trying to catch the state of primary arches.
I don't think that we need a 31bit installer, the latest
hardware is already several years old.
On the other hand, the kernel and the userspace are built to run only on
z9 or better (z10) hardware, so having an option for users of eg. z900
makes sense. Even if it would be presented with a root filesystem image
in the first phase.