Re: Unidentified subject!
Christian T. Steigies wrote:
> > >
> > > | > If in fact I, as maintainer, have a choice in the matter , then I would like
> > > | > to request the same for the following packages:
> > > | >
> > > | > octave2.1, quantlib, r-base
> > > | >
> > > | > for the
> > > | >
> > > | > arm, m68k
> > > | >
> > > | > architectures. I have spent *way* too much fscking special requests for
> > > | > these smaller + older architecture which are, quite simply, mismatched for
> > > | > these numerically-focussed applications and environments.
> > > |
> > > | Please convince a porter for those architectures to add the right lines
> > >
> > >
> > > How would I do that? Whenever I discuss this with people from the porting
> > > teams, their attitude usually is "why -- we may as well build it". Which is
> > > wrong, IMHO, as these arches _do_ hold up releases of these packages more
> > > often than I like.
> > >
> > > I still need help in this matter.
> > >
> > > Regards, Dirk
> > >
> > > | to Packages-arch-specific. If I'd remove those now, they'll get built
> > > | and re-uploaded immediately again, so that has no use.
> > > |
> > > | Update this bug if the changes to P-a-s are done.
> > > |
> > > | --Jeroen
> >
> > octave2.1, quantlib, r-base added to P-a-s for m68k. Request
> > of maintainer (for whatever reason) is always good enough to
> > have a package removed.
>
> I thought this would have been solved by making most arches and the porters
> second class citizens, i.e. releases should not be stopped by "minor" arches
> anymore. But I haven't followed the Vancouver discussion during the last
> month, has it been decided that kicking out individual packages, which run
> just fine on "minor" arches, is the way to go now?
>
> By not building octave and r-base anymore, dozens of packages will become
> unbuildable on m68k (and arm), do you have to add them to P-a-s, too, so that
> those can move into testing?
There's more info in with the bug, #167780. Dirk says:
> I have a lot of respect for the m68k crew, but the fact remains that these
> two arches make my life a lot harder, place a burden on regular users on
> i386, powerpc, alpha, ... without _any_ benefit to m68k or arm users as
> there simply aren't any for these apps.
It's really the maintainer's call if he's not willing to
support an arch. No one else is doing any meaningful
porting work.
It does affect other packages. I'll see how tangled it is.
P-a-s was munged up earlier and I didn't add these 3, although
I've added them to nfu.
I guess we'd better talk some more about this before doing
anything. I'll start another thread in the m68k groups.
Rick
--
Reply to: