Describe the issue
When do you need this? (YYYY/MM/DD)
When is this no longer needed or useful? (YYYY/MM/DD)
If we cannot complete your request, what is the impact?
Please note that we are not asking a new mass rebuild, just that the current mass rebuild continue as planned.
Notes:
Early testing of GCC 14 has started using mass-prebuild build orchestration via COPR to test the impact on the distribution and to verify that the rebase to the new gcc will go smoothly.
glibc has been doing weekly rawhide rebases to upstream for the last 6 months as usual and fixing integration issues as they arise.
binutils has been through continuous testing in rawhide since August 2023 and no integration issues have been detected.
Metadata Update from @phsmoura: - Issue tagged with: changes, f40
Hello @codonell,
Thanks for opening this ticket, the mass rebuild is scheduled for Wed 2024-01-17. I hope everything is on track, do you want us to inform you on any internal channels (if yes please point at those) I'll be happy to assist.
Please let us know if there is anything else on the releng end!
The Fedora teams supporting the GNU Toolchain met on January 10th and at that time we were ready to go with the mass rebuild planned for January 17th.
As of Saturday January 13th the new GCC 14 is in the repo, and built into a side tag: https://koji.fedoraproject.org/koji/buildinfo?buildID=2344783
glibc and binutils are built and available already in Rawhide.
The f40-build-side-81394 side-tag now contains gcc, annobin, libtool and various Ada packages. We are waiting with tagging that into rawhide on the acceptance of the https://fedoraproject.org/wiki/Changes/GNUToolchainF40 change, but unfortunately even when it has been announced 19 days ago, it doesn't seem to have a ticket filed.
The Port to Modern C change request (https://fedoraproject.org/wiki/Changes/PortingToModernC) is already accepted for F40 and we need to merge GCC 14 to avoid regressing those changes.
My position is that FESCo's acceptance of the Port to Modern C change is tacit acceptance that we will keep moving forward with our annual rebase of the GNU Toolchain.
Since Fedora 35 we have been submitting twice-yearly GNU Toolchain rebase system-wide change requests as a mechanism for improving our collaboration with the whole of the Fedora community. This year the side-tag build collaboration has been even better than previous years and we can keep improving that over time.
My opinion is that the GCC 14 side tag should be merged today regardless of the current acceptance status of the GNUToolchainF40. The technical risk of not having GCC 14 in the mass rebuild is too high.
The newer binutils and newer glibc are already in place and being used.
I would like to add that as an improvement this year we have all of the Ada packages built into the side-tag thanks to Björn Persson and so the situation that happened last year with the late-building of Ada packages is avoided. I'm always excited to see the system-wide change request process result in positive impact and smoother Fedora updates.
The side tag is merged: https://bodhi.fedoraproject.org/updates/FEDORA-2024-286f139579
FESCo process was handled here: https://pagure.io/fesco/issue/3145
I don't think there is anything left for release engineering at this point and we've resolved the issues surrounding the system-wide change.
Hello,
Thanks for opening this ticket and looping us in, there were some releng issues opened against this I guess, which were taken care of, and I don't think we have anything else left for this. Closing this, but feel free to open this in case of any issues.
Metadata Update from @jnsamyak: - Issue close_status updated to: Fixed - Issue status updated to: Closed (was: Open)
Log in to comment on this ticket.