Monday Patch Queue Review update (2026-06-29)

[Carlos O'Donell]

Most recent meeting status is always here:
https://sourceware.org/glibc/wiki/PatchworkReviewMeetings#Update

Meeting: 2026-06-29 @ 0900h EST5EDT

Video/Audio: https://bbb.linuxfoundation.org/room/adm-alk-1uu-7fu

IRC: #glibc on OFTC.

​ * State NEW delegate NOBODY at 633 patches.
  * Starting review of release blockers and desirables.
  * s390: Prevent hoisting the thread-pointer read in THREAD_SET_STACK_GUARD (BZ 34297) (Adhemerval)
   * Reviewer Pool: Carlos O'Donell
  * Wilco: Rename vector math functions?
  * Adhemerval: Reviewed v2 and installed.
  * posix: Fix wordexp WRDE_APPEND to preserve state on non-NOSPACE errors (BZ 34090, CVE-2026-6368) (Adhemerval)
   * Reviewer Pool: Florian Weimer
  * Discussed simple bug fixes and doc changes are OK for contributors participating in Arjun's DevConf talk/workshop.
  * Discussed gdb defect for the malloc IFUNC changes.
   * Do not revert glibc changes.
   * gdb changes are well on the way.
   * Florian: Do we need to work around this in glibc?
   * Yury: Can open a gdb bug with a patch to disable the functionality.
    * Carlos: Updated wiki with packaging changes with notes about gdb 17.2.
    * Adhemerval: Why haven't we seen this before?
    * Yury: gdb calls the inferior's malloc to allocate memory, and it didn't consider ifunc.
  * Going through "Deisrable this release?"
  * TCB layout changes:
   * Florian: We believe the fixes have been rolled out. Shall we test this in Rawhide?
   * Florian will put these into Rawhide and report back results.
  * Discussed "elf: Make string tunables startup-only"
   * Adhemerval: We can defer this to the next release.
   * Carlos: Please note parse_tunables_string() fails to return non-zero when it finds tunables. Would be nice to fix in a series that cleans this up.
  * elf: Remove dead l_need_tls_init static-TLS init path (Adhemerval)
   * Reviewer Pool: Florian Weimer
  * system-wide tunables
   * Arjun reviewing for DJ on v9 and should be close to inclusion.
   * Exiting in-progress feature that we would allow in the soft-freeze period.
  * Improve make and make check parallelism (Adhemerval)
   * Under review from Sam James (distro).

-- 
Cheers,
Carlos.