Monday Patch Queue Review update (2023-05-15)

Monday Patch Queue Review update (2023-05-15)

[Carlos O'Donell]

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

Meeting: 2023-05-15 @ 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 277 patches.
 * Carlos' SLI at ~154 days average and 48253 total patch days.
 * Starting at 69335.
 * x86_64-gnu signal fixes (Sergey)
  * Hurd maintainers to review.
 * [v9,1/3] x86: Increase `non_temporal_threshold` to roughly `sizeof_L3 / 4` (Noah)
  * Carlos and DJ will review the patches.
 * Patch to fix glibc condition variable bug (BZ 25847) (Malte)
  * Carlos is writing open-box tests for this and reviewing.
 * [v2] S390: Use compile-only instead of also link-tests in configure. (Stefan)
  * Accepted by Andreas.
 * powerpc:GCC(<10) doesn't allow -mlong-double-64 after -mabi=ieeelongdouble (Mahesh)
  * Florain to post a response.
 * MIPS: Sync elf.h from binutils (Ying)
  * Andreas requested changes.
 * time: strftime_l: Use malloc rather than an unbounded alloca. (Joe)
  * Needs review of the malloc size computation.
 * setsourcefilter: Use malloc() rather than alloca(). (Joe)
  * Needs review of the malloc size computation.
 * [v2,1/2] elf: Fix marking root dir as nonexist in open_path (bug 30435) (Qixing)
  * Carlos to review.
 * [v1] x86: Use 64MB as nt-store threshold if no cacheinfo [BZ #30429] (Noah)
  * Florian to follow up with a comment.
 * [v4] Restore support for _r_debug copy relocations & interposition (Florian)
  * Needs review from someone with ld.so experience.
 * Remove last remnants of have-protected (Andreas)
  * Carlos to review.
 * [v4,1/3] posix: add (failing) test for REG_STARTEND (Ahelenia)
  * Still needs review of test case and gnulib sync status.
 * [v12] POSIX locale covers every byte [BZ# 29511] (Ahelenia)
  * Still with Florian for review.
 * [v6] x86_64: aarch64: Set call number just before syscall (Joe)
  * Joe should ping HJ for review of the x86_64 changes.
  * Split this into 3 patches so we can make forward progress.
  * Adhemerval to ask Joe to update "NewPorts" wiki page with QoI requirement.
 * Stopped at 68153.
 * Florian asked Palmer if there was a reproducer for the case of dlopen'ing executables
  * Palmer: If gp was used by a shared object, then that is likely a bug in ld.
  * Palmer: Reusing gp for other purposes is a violation of the psabi.
  * More details on what they are doing would help, including a reproducer.
 * Adhemerval: Florian only reviewed the last of the functions for pid fd.
  * Florian will try to look at this during the week.
 * Carlos: Where did we land with the huge pages on stack allocation?
  * Adhemerval: Installed Cupertino's patch to add the tunable.
  * Posted new patch to try detect the bad performance case and fix it.

-- 
Cheers,
Carlos.