Better late than never, glibc 2.41 should land in -proposed in during the
weekend. While upstream has done its usual job, I usually "bake" the
release much more than this, so there might be a bit more unforeseen
breakage than usual this time around, my apologies.
A high-level overview of the changes is available at
https://lists.gnu.org/archive/html/info-gnu/2025-01/msg00014.html
The good news is packages that do not use the new APIs shouldn't get
stuck behind the glibc transition, as symbols versioned to 2.41 didn't
exist beforehand.
The following are semi-educated guesses as to what could pose some
issues:
* getenv/setenv, if you're doing weird things with it
(as for instance bash), since they've seen significant work regarding
thread safety
* floating point calculations on ppc64el: a few tests are failing on
NaNs on ppc64el. I doubt it'll break real-world code but I have been
known to be wrong once in a while.
* execstack-related failures: https://sourceware.org/bugzilla/show_bug.cgi?id=32653
Cheers,
Simon
--
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
