--- Begin Message ---
- To: Paul Gevers <elbrus@debian.org>, 1026199-done@bugs.debian.org
- Subject: Re: Bug#1026199: release.debian.org: Is the toolchain list updated for bookworm
- From: Sebastian Ramacher <sramacher@debian.org>
- Date: Fri, 11 Apr 2025 21:09:01 +0200
- Message-id: <Z_lozSBS0gAhEig0@ramacher.at>
- In-reply-to: <313327ea-ca05-6ec9-b25f-3637fe139947@debian.org>
- References: <167115482282.292444.17330970584235210621.reportbug@industrial-algebra.suchdamage.org> <167115482282.292444.17330970584235210621.reportbug@industrial-algebra.suchdamage.org> <f8fc35f6-443a-f979-3702-dc86dfffbb2f@debian.org> <167115482282.292444.17330970584235210621.reportbug@industrial-algebra.suchdamage.org> <Y9Dsvy8/XpsBl7cV@thunder.hadrons.org> <Y/o8ys6c/9HVnyMv@thunder.hadrons.org> <167115482282.292444.17330970584235210621.reportbug@industrial-algebra.suchdamage.org> <313327ea-ca05-6ec9-b25f-3637fe139947@debian.org>
On 2023-03-02 13:17:36 +0100, Paul Gevers wrote:
> Hi,
>
> Sorry for the delayed answer, also to Sam's original question.
>
> On 25-02-2023 17:52, Guillem Jover wrote:
> > On Wed, 2023-01-25 at 09:47:59 +0100, Guillem Jover wrote:
> > > Ah, was wondering the same few days before the toolchain freeze, as I
> > > was unsure whether to update some of the packages I maintain (in
> > > particular libmd, for which I was thinking of doing a new upstream
> > > release), and was also checking hints for explicit blocks. See below…
> >
> > Could we get an answer to the below question? I'm currently uncertain
> > how to proceed with libmd as I have pending to do an upstream release
> > with targeted fixes, and have several packaging fixes too. And
> > depending on the answer below I'd either discard one or both, and
> > request or not a pre-approval unblock, or forget about it. I've also
> > just pondered simply requesting a pre-approval unblock, given that to
> > me libmd seems clearly to be in the essential-set now, but… :)
> >
> > (I don't greatly mind the answer, say "yes, they should be but because
> > they were not included we will include them on the next freeze", or "yes,
> > they should be considered", or "no, they are not", or anything else, I'm
> > more interested on knowing how to proceed to get this off the back of
> > my head. :)
>
> I would hope that [1] describes what we're looking for with this list:
> """
> Debian has quite some packages that influence the content of packages and
> how they are built, i.e. most packages that are part of (build-)essential
> and other toolchain packages. The impact of bugs in these packages can take
> a lot of time to resolve because it requires figuring out which packages are
> affected and how they need to be rebuild after the bug itself is fixed.
> Therefore, changes to these kind of packages are no longer appropriate.
> """
>
> The policy goes on with saying that that definition is (slightly) vague.
> What we are trying to avoid is changes to packages that influence the
> content of (lots of) binaries when build. I'm now realizing that our
> hard-coded list might be including a bit more (mostly libraries) packages
> than we really mean, but from our perspective that would be erring on the
> safe side and I think quite some work to understand how each of these is
> influencing the build in order to remove them from that list.
>
> And yes, pre-depends was missing from the list; when I add it, libmd0 shows
> up. I'll hopefully update the list shortly (after a bit more checking on the
> additions and the MR).
>
> As with all our rules, there's room for exceptions (decreasing while we
> progress towards the release), so please show the proposed changes with
> taking all of the above in mind (and maybe after reading our unblock FAQ
> [2]) if you believe you can justify the upload.
>
> [1] https://release.debian.org/testing/freeze_policy.html#transition
> [2] https://release.debian.org/testing/FAQ.html
I think we can close this bug. bookworm was released and we have updated
the list for trixie. If there is a need to update the list of toolchain
packages for trixie, please open a new bug.
Cheers
--
Sebastian Ramacher
--- End Message ---