Monday Patch Queue Review update (2024-01-22)

Monday Patch Queue Review update (2024-01-22)

[Carlos O'Donell]

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

Meeting: 2024-01-22 @ 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 461 patches.
 * Carlos's SLI at 260 days average patch age in queue and 121140 accumulated patch days.
 * Patch review is looking at glibc 2.39 release blockers and desirable.
 * No blockers for today.
 * No desirable patches for today.
 * Carlos: Note Fedora 40 downstream mass rebuild happened last week with GCC 14, Binutils 2.41 and glibc 2.39 intermediate build with updated qsort changes from upstream as of e2803cfd8bb00b54816d1a2c381d2cf82b136da6. Results looked good overall.
 * Adhemerval: Static pie patch for RISC-V needed additional work.
  * https://patchwork.sourceware.org/project/glibc/patch/mvmplxzvv2j.fsf@suse.de/
 * string: Disable stack protector for memset in early static initialization
  * fweimer: Should fix the generic implementation.
  * Andreas: +1 from RM.
 * Carlos: Florian, anything that we saw in the Fedora 40 changes?
  * Florian: Yes, downstream-only binutils bug, unrelated to glibc.
 * Florian: 2 NEWS and 1 manual update.
  * Florian: Need to repost one patch changes.
  * Florian: Forgot the ABI change for utmp and utmpx, will propose for glibc 2.40.
  * Adhemerval: Do you suggest we keep the interface in the next release?
  * Florian: I don't know, we probably need to disable it in the headers.
  * Florian: We will make the change downstream so we don't run into problems.
  * Adhemerval: SUSE did a lot of work to fix this and we should go in that direction.
 * Szabolcs: Found a BE-only vector math code bug for AArch64 but we can fix it and backport.
 * Carlos: We should have a script that captures the results? Then devs could just paste the results.
 * Carlos: The nptl/tst-thread-exit-clobber is odd, but ppc32 and ppc64 is an issue that Raji has to review.
 * Florian: Note that if you run with the python ABNF parser then it might timeout tst-rtld-list-diagnostics.
 * Andreas: Noted that test-cxa_atexit-race2 was a timeout for s390x so we don't have to worry there.
 * Florian: We should add a new test exit code for timeout to make it clear in the reporting.
  * Carlos: Could still be a timeout due to deadlock, but knowing that status is better than not knowing.
 * Carlos: I need to do machine testing this week and add notes to the wiki. I will look at a script to collect the items.
 * Carlos: I want to reivew this https://patchwork.sourceware.org/project/glibc/patch/779c4fcc-5dc9-4bc1-9d68-cbce31a0d8e7@redhat.com/ and include it in the release. I'll TO the RM and get approval.

-- 
Cheers,
Carlos.