Hi, On 2024-07-14 19:57, Sebastian Ramacher wrote: > Control: tags -1 confirmed > Control: forwarded -1 > https://release.debian.org/transitions/html/glibc-2.39.html > > On 2024-07-08 07:26:43 +0200, Aurelien Jarno wrote: > > Package: release.debian.org > > Severity: normal > > X-Debbugs-Cc: gl...@packages.debian.org > > Control: affects -1 + src:glibc > > User: release.debian....@packages.debian.org > > Usertags: transition > > > > Dear release team, > > > > I would like to get a transition slot for glibc 2.39. It has been > > available in experimental for two months already. It has been built > > successfully on all release architectures and most ports architectures. > > The experimental pseudo-excuses look good overall. > > Please go ahead.
Thanks, I have just uploaded it. > > The current known issues are available in the BTS using the glibc2.39 > > usertag: > > https://bugs.debian.org/cgi-bin/pkgreport.cgi?tag=glibc2.39;users=debian-gl...@lists.debian.org > > > > gopacket has a patch available and I can take care of NMUing if it is > > not fixed before the transition starts. > > > > For aspectc++, cbmc and rocm-hipamd, the situation is a bit more > > complex. Those packages have issues with the types introduced on the > > arm64 version of bits/math-vector.h. Those are guarded by clang or gcc > > version checks, but the guards are ignored by the packages for various > > reasons. A workaround is present in glibc 2.38, but it can't be ported > > easily in glibc 2.39. I therefore propose to remove the corresponding > > arm64 packages from the archive. aspectc++ and cbmc are leaf packages. > > For rocm-hipamd, this also means removing 15 reverse dependencies. > > We can wait a bit for the maintainers to react and otherwise let's go > ahead with the removals. Ok, I have upgraded them to serious. Note that at least for gopacket and rocm-hipamd, these bugs translate into autopkgtest regressions. Regards Aurelien -- Aurelien Jarno GPG: 4096R/1DDD8C9B aurel...@aurel32.net http://aurel32.net
