rawhide report: 20090408 changes

[Orcan Ogetbil]

On Wed, Apr 8, 2009 at 12:49 PM, David Nielsen wrote:
>
>
> 2009/4/8 Orcan Ogetbil <oget.fedora at gmail.com>
>>
>> On Wed, Apr 8, 2009 at 11:16 AM, Peter Robinson wrote:
>> >>>>> Broken deps for ppc
>> >>>>> ----------------------------------------------------------
>> >>>>>       avahi-sharp-0.6.24-2.fc11.ppc requires mono(System) =
>> >>>>> 0:1.0.5000.0
>> >>>>
>> >>>> [cut]
>> >>>>>
>> >>>>>       taglib-sharp-2.0.3.2-2.fc11.ppc requires mono(mscorlib) =
>> >>>>> 0:2.0.0.0
>> >>>>>       tasque-0.1.8-2.fc11.ppc requires mono-core
>> >>>>>       tasque-0.1.8-2.fc11.ppc requires mono-data-sqlite
>> >>>>>       themonospot-0.7.1.1-2.fc11.ppc requires mono-core >= 0:1.2.3
>> >>>>>       webkit-sharp-0.2-1.fc11.ppc requires mono(mscorlib) =
>> >>>>> 0:1.0.5000.0
>> >>>>>
>> >>>>
>> >>>> Why is there a huge number of broken mono dependencies, only in ppc.
>> >>>> Did we miss something?
>> >>>
>> >>> Read further up the mail where is states that mono on ppc has been
>> >>> disabled.
>> >>>
>> >>> Peter
>> >>
>> >> But first of all, such change (i.e. disabling ppc support on mono
>> >> package)
>> >> must not be done at this stage where F-11 final freeze comes very soon
>> >> (4/14).
>> >> If it cannot be done to fix the latest mono package to support ppc
>> >> again,
>> >> the lastest mono package must be untagged.
>> >
>> > There has been a number of discussions about this on fedora-devel over
>> > the last week or so, check the archives for details.
>> >
>>
>> What I don't understand is, shouldn't such a change that affects
>> dozens of packages go through FESCo? If it did, why did we not see an
>> announcement with BIG CAPITAL LETTERS with cream on top?
>
> Not a FESco decision by a mile, it is a bug... patches welcome as always.
> You also were warned on -devel, several times in fact that we had this
> problem. _Nobody_ with ppc skills volunteered to help actually fix the
> problem so now it's disabled till the ppc arch team helps out with a
> solution. Nobody on the Mono SIG has the hardware to test this nor the in
> depth knowledge about the architecture to fix it. It has been reported to
> upstream as well so hopefully we will see this situation mended soon.
>
> - David
>
>
>

Ah... I give up. You guys are right, by doing this update which breaks
many packages in an architecture. I'm wrong.
Sorry,

Orcan