Re: which kernel version for etch?
On Wed, May 10, 2006 at 10:15:45AM +0200, Bastian Blank wrote:
> On Wed, May 10, 2006 at 07:46:12AM +0200, Sven Luther wrote:
> > > - branch the current linux-2.6 package into a new source package, say
> > > linux-2.6.16, tracking 2.6.16.x releases
> > Sounds the right thing to do for me. We keep linux-2.6 as trunk package, and
> > can branch off any number of kernel we want to have around long term, and mark
> > it as linux-2.6.<x>. Problem is with the metapackages, where will they come
> > from ? Bastian, can you comment on this ?
>
> Currently they are built from linux-2.6. This needs to be changed for
> such a package. And after that, they need to be reintroduced through
> t-p-u.
Well, this means thought that they will no more be built with linux-2.6, right
?
Why do you need to involve t-p-u, the current version in testing points to the
linux-2.6 2.6.16-<something> which is compatible with the new packages. We
just upload linux-2.6.16 to unstable and linux-2.6 without the metapackages,
and they should migrate to testing normally or with an RM hint at the right
moment, no ?
> > I hearthily vote for this, and my effort to produce the .udebs from the common
> > kernel and get a solution for the out-of-tree modules where designed to allow
> > this easily enough, even in case of abi-changes. Support for those ideas have
> > been tentative at best, especially in the light of the extreme flamming i got
> > from the d-i folk about this.
>
> Just provide them and either they want them (at least for s390, I do)
> or they don't.
Ok. This means that at least 2 architectures, powerpc and s390, are interested
in this. Thanks for your support in this, i will see what we can do, probably
in a branch for now until the situation is proven. I am little interested in
going this way, if it means another flamewar between me and the d-i team, with
everyone else watching silently.
Andreas Barth, you claimed that you could see strong reason not to go this
way, please could you comment on them now in this thread.
> For the out-of-tree modules, we will see it in the next days, if it
> works. I think this should be done in at most 4 days. If it does not
> work, we have to take down the law and build them ourself.
Documentation is needed here also.
> > We would also need to get some policy fixed about what happens to user
> > self-compiled modules in that case.
>
> Either the ABI is compatible or not.
A, sure, that is hardly the problem, what we need is some documentation, which
will inform the user of what he is supposed to do in case of ABI changes.
Friendly,
Sven Luther
Reply to: