On 2025-01-30 17:33, Kevin Fenzi wrote:
On Wed, Jan 29, 2025 at 10:01:09AM +0100, Vít Ondruch wrote:
Thanks for clarifying. Of course one reason for mass rebuild can be that we
are not able to properly identify the package set for more targeted mini
mass rebuild.
But IMHO, having just the Copr build to identify the problematic packages is
good enough. If we want more, e.g. if new GCC brings some performance
optimizations, rebuilding just the 5001 specifically for this occasion and
the rest done later by mass rebuild e.g. after branching, that would be
still fine. If we want the rebuild due to e.g. glibc changes, we sill can
come up with more targeted package set.
Right, so you are advocating for mass rebuilds to always be targeted to
only the needed subset? I guess I think thats pretty nice, but I also
think it could mean a lot of work for releng if they have to identify
those packages. Perhaps we could move to requiring change owners to
identify all packages they want to have rebuilt?
mass prebuilder kinda does this already using rpm dependencies; that's
what we use to identify and rebuild packages for changes to individual
toolchain components.
Maybe if the mass prebuilder checkpoint gets accepted and we're able to
stabilize that, we could use that for canonical rebuilds (instead of
just scratch build tests) in place of the January mass rebuild?
Sid
--
_______________________________________________
devel mailing list -- devel@lists.fedoraproject.org
To unsubscribe send an email to devel-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
Do not reply to spam, report it:
https://pagure.io/fedora-infrastructure/new_issue
