13:01:49 <jki> #startmeeting CIP IRC weekly meeting 13:01:49 <collab-meetbot> Meeting started Thu Apr 17 13:01:49 2025 UTC and is due to finish in 60 minutes. The chair is jki. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:01:49 <collab-meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:01:49 <collab-meetbot> The meeting name has been set to 'cip_irc_weekly_meeting' 13:01:58 <jki> #topic AI review 13:02:02 <jki> - check release scripts/procedures shared by iwamatsu-san [pavel] 13:02:35 <pave1> I took a look. They'd need a bit of docs and quite a lot of documenting prerequisites. 13:03:14 <pave1> Plus, trying to modify someone else's shell code ... is not easy. 13:04:02 <iwamatsu__> pave1: OK, I will create a document. 13:04:41 <pave1> Mainly directory tree it expects to work in would be useful. 13:06:07 <jki> ok, then there is clear next step 13:06:37 <jki> further concrete wisches/feedback may also help writing the doc ;) 13:06:51 <jki> other AIs? 13:07:01 <jki> 5 13:07:03 <jki> 4 13:07:05 <jki> 3 13:07:06 <jki> 2 13:07:08 <jki> 1 13:07:10 <jki> #topic Kernel maintenance updates 13:07:26 <masami> This week reported 128 new CVEs and 40 updated CVEs. 13:07:28 <uli_> i'm working on 4.4 13:07:34 <pave1> I was reviewing 6.1.134 with a bit of .132 and AUTOSEL. 13:07:44 <pave1> uli: That gcc crash looks like a lot of fun. 13:07:59 <pave1> uli: My guess would be that something in headers changed and now trips the compiler bug. 13:08:26 <uli_> and i absolutely cannot reproduce it, even with the same compiler version. 13:08:42 <iwamatsu__> I reviewed 6.1.134 and reviewing .132 13:09:20 <uli_> any suggestions on how to proceed with that? the builds that succeed also work... 13:09:28 <pave1> uli: It probably counts as a security issue in gcc... so fun. May depend on CPU architecture, for example. 13:09:47 <uli_> iirc it only happens on arm. 13:09:57 <pave1> uli: So it works locally but breaks on gitlab? 13:10:20 <uli_> yes. some configs don't build, some do. 13:10:29 <uli_> those that build also pass the tests 13:11:07 <pave1> Easist thing would be to find out what change trips that gcc bug and just not apply that one... 13:11:31 <jki> then let's first try to update the compile stable release in the breaking environment 13:11:55 <uli_> +1. that's a lot less work, too... 13:12:13 <jki> who can do that? 13:12:43 <uli_> patersonc, i would assume. 13:13:31 <jki> ask him via the list - but he might be on vacation these weeks 13:13:53 <uli_> will do. 13:14:43 <jki> some fun from the -rt corner, in case you didn't read this yet: 13:14:46 <jki> https://lore.kernel.org/all/20250409120746.635476-1-ziqianlu@bytedance.com/ 13:15:05 <jki> high latency for non-rt, RCU stalls for -rt (all versions) 13:15:47 <jki> once the series is finished, we will also ask for backports - but < 6.12 is... different 13:17:13 <jki> it's one of those "don't use container features on RT" pavel warned about :) 13:17:31 <jki> anything else? 13:17:55 <jki> 5 13:17:56 <pave1> Yeah. RT is hard. Containers are hard. 13:17:59 <jki> 4 13:18:02 <jki> 3 13:18:04 <jki> 2 13:18:06 <jki> 1 13:18:09 <jki> #topic Kernel release status 13:18:15 <jki> all on track 13:18:40 <jki> but some are due very soon 13:18:46 <jki> any blockers in sight? 13:18:47 <pave1> Yep, 4.4-rt. 13:19:03 <pave1> Nothing unusual. 13:19:05 <iwamatsu__> I am going to release 5.10 and 6.1 tomorrow. 13:19:09 <jki> perfect 13:19:19 <jki> then let's move on 13:19:22 <jki> 5 13:19:23 <jki> 4 13:19:25 <jki> 3 13:19:27 <jki> 2 13:19:29 <jki> 1 13:19:30 <jki> #topic Kernel testing 13:19:46 <jki> probably not much news here today... 13:20:18 <arisut> currently testing config that I'm moving to KernelCI 13:21:22 <arisut> with kci_build kernelci internal command 13:22:33 <arisut> that's all 13:22:55 <jki> ok, thanks 13:23:02 <jki> anyone anything to add? 13:23:31 <jki> 5 13:23:33 <jki> 4 13:23:34 <jki> 3 13:23:36 <jki> 2 13:23:37 <jki> 1 13:23:39 <jki> #topic AOB 13:23:49 <pave1> 8 13:24:53 <jki> yes? 13:25:23 <jki> further topics for today? 13:25:25 <pave1> No, sorry. Stray typo. 13:25:50 <jki> 5 13:25:51 <jki> 4 13:25:53 <jki> 3 13:25:55 <jki> 2 13:25:57 <jki> 1 13:25:58 <jki> #endmeeting