On Tue, 2002-09-17 at 15:26, Matt Reynolds wrote: > On Fri, 2002-09-13 at 15:55, David Pashley wrote: > > KDE 3.0.3 is packaged and apart from a KDE 3.0.4 release, the KDE 3.0.x > > branch is pretty much finished packaging wise. calc and others are now > > working on KDE 3.1 packaging. > > > > KDE 3.x will be in sid once the GCC transition plan has been worked out > > and started. This is kind of out of our hands. > > I understand that the GCC transition is not under your control, and > please understand that my criticism applies to the entire process, not > just the KDE packaging. I would love to see more information concerning > the GCC process as well. I have read the technical material concerning > the transition, but a laymens, 20000 foot view, with approximate dates > and places to go find new information to track the process, would be > wonderful.
AFAIK there's no such place, currently. I read debian-gcc and
debian-glibc because I'm mildly interested (in a i-use-only-x86-for-now-
but-it's-cool-to-have-an-OS-that-runs-on-11-arches way), and from what I
can see these are the issues right now:
!!! everything IIRC, I'm not involved personally !!!
- gcc 3.2 and glibc 2.3 Gang Together
- gcc code generator is still not 100% stable on all platforms
(ARM is one of the problems, binutils update might be
required)
- glibc still needs tweakings, here HPPA is one of the problems.
for both, m68k is problematic if only because cross-building is
not
supported for debian pkgs, and the native builds takes AGES.
- perl 5.8 creates some problems with the build system now and then.
No idea on the timeline, but the work Will Be Done (tm).
While bundling glibc and gcc is a good thing imho, it slows down the gcc
release which would probably be ready earlier if done with the current
glibc.
(someone involved correct me please if I got any facts wrong here).
cheers
-- vbi
--
secure email with gpg http://fortytwo.ch/gpg
NOTICE: subkey signature! request key 92082481 from keyserver.kjsl.com
signature.asc
Description: This is a digitally signed message part
