Monday Patch Queue Review update (2025-03-24)

[Carlos O'Donell]

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

Meeting: 2025-03-24 @ 0900h EST5EDT

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

IRC: #glibc on OFTC.

Review new patches and restart review at the top.

  * State NEW delegate NOBODY at 509 patches.
  * Carlos's SLI at 429 days average patch age in queue and 224744 accumulated patch days.
  * Starting at 108937
  * [v3] mach: Use the host_get_time64 to replace the deprecated host_get_time for CLOCK_REALTIME when … (Luo)
   * Needs Hurd maintainer review.
  * [v3] linux: Add openat2 (BZ 31664) (Adhemerval)
   * Adhemerval: Comments from v1 were from Florian.
   * Carlos: What you have seems sensible.
   * Reviewer Pool: Arjun Shankar.
  * stdlib: Consolidate getentropy and adapt to POSIX 2024 semantics (Adhemerval)
   * Adhemerval: Recent linux kernels always provide enough entropy.
   * Carlos: Reviewer question: Do we version this for the EIO vs. EINVAL change?
   * Reviewer Pool: Carlos O'Donell
  * malloc: Use __always_inline for simple functions (Wilco)
   * Reviewer Pool: DJ Delorie
  * malloc: Use _int_free_chunk for remainders (Wilco)
   * Reviewer Pool: DJ Delorie
  * Linux: Inhibit tail calls in cancellable system calls (fweimer)
   * Discussed the impact to other architectures.
   * Adhemerval: For some architectures this results in another call.
   * Florian: Users may expect to see the syscall in the listed call frame, and with the current linux syscall ABI the syscall number if often lost right away and not saved anywhere so you don't know the syscall. It may be the only way to identify the syscall.
  * Stopping at 108688.
  * Yury: There is some conflict between memory sealing and BTI.
   * Adhemerval: A binutils bug where it was generating wrong PLT for standard branch protection.
   * Yury: When we want to link a binary with both memory sealing and BTI the linker checks the markings in other objects, then it should use the correct sequences for the PLT. If at the same time we request memory sealing it uses the incorrect sequence for the PLT.

-- 
Cheers,
Carlos.