What's left in unreleased.
Hi,
currently, these packages are still in d-p.o unreleased:
- eglibc
Submitted, they said they will upload after squeeze is out.
- elfutils
Needs porting. If someone feels like doing it, please do so.
Otherwise, wait until I get bored enough ;-)
- gcj-4.4
This doesn't need to be there, the version from unstable can
be built just fine, used too, except the 4.4.5-9 one we had
there would need a binNMU against eglibc_2.11.2-7+m68k.4 to
work in a useful way. So ignore it; the next upload of this
source package to unstable will move this out.
- gmp
Fixed in experimental. Hah. Someone other than me should
prod maintainer about uploading a fixed version to unstable
as well. (Just in case the one from experimental breaks…)
The fix is a two-liner.
- gobject-introspection
I was asked whether the version from experimental exhibits
the same bug (alignment/padding assumptions wrong), but
couldn't build it due to b-d not up to date. Maintainer
seems reluctant to fix it in unstable. However, it's just
an unimportant b-d for udev which installs without it.
- linux-2.6
- linux-latest-2.6
These should be obvious. We should try to work on the kernel
from experimental. I have begun preparing for that before I
left for VAC, but have yet to pick it up again. Maybe, when
I manage to build a kernel based on it with the m68k-queue
patches applied, this counts as test of those patches, and
they might be pushed? More people could do that IMHO. One
linux-2.6 build takes me about five days.
- m68k-missing
Transitional package. This used to be just enough glue for
locales and to be able to use linux-libc-dev (instead of
linux-kernel-headers which got removed) together with the
old glibc 2.7 we used to have in unstable. Nowadays, this
package is empty and only depends on the new TLS eglibc.
I will ask Aurélien to remove it by end of 2011 since all
systems hopefully are up to date by then, and not too many
people would have gotten the old package anyway.
That’s about it. Meaning, after eglibc, bootstrapping from
unstable might be possible again, although --variant=buildd
probably needs gmp and possibly linux-kernel-headers. For
now, there’s still my repo, being updated constantly, and
the tarball I published, which will be updated after the
next round of compiles.
bye,
//mirabilos
--
“Having a smoking section in a restaurant is like having
a peeing section in a swimming pool.”
-- Edward Burr
Reply to: