12:00:32 <jki> #startmeeting CIP IRC weekly meeting
12:00:32 <collab-meetbot> Meeting started Thu Jun 25 12:00:32 2026 UTC and is due to finish in 60 minutes.  The chair is jki. Information about MeetBot at http://wiki.debian.org/MeetBot.
12:00:32 <collab-meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
12:00:32 <collab-meetbot> The meeting name has been set to 'cip_irc_weekly_meeting'
12:00:39 <jki> #topic AI review
12:00:44 <jki> none on the list
12:00:50 <jki> 5
12:00:52 <jki> 4
12:00:54 <jki> 3
12:00:55 <jki> 2
12:00:56 <jki> 1
12:00:58 <jki> #topic Kernel maintenance updates
12:01:10 <masami> This week reported 223 new CVEs and 104 updated CVEs.
12:01:27 <uli> i'm still working on 4.19
12:01:29 <iwamatsu> I reviewed 6.12.91, and  I am reviewing .92.
12:01:37 <pave1> I was doing reviews: 6.12.94.
12:02:41 <jki> anything to add?
12:02:51 <jki> 5
12:02:54 <jki> 4
12:02:55 <jki> 3
12:02:57 <jki> 2
12:02:59 <jki> 1
12:03:01 <jki> #topic Kernel release status
12:03:11 <jki> just 6.12 is due
12:03:33 <iwamatsu> I am going to relelase tomorrow.
12:03:43 <jki> great
12:03:55 <jki> anything else on releases?
12:04:03 <jki> 5
12:04:06 <jki> 4
12:04:08 <jki> 3
12:04:10 <jki> 2
12:04:12 <jki> 1
12:04:14 <jki> #topic Kernel testing
12:05:14 <patersonc> No updates from me
12:05:15 <jki> anything regarding testing?
12:05:23 <jki> ok
12:06:15 <jki> still device troubles in Munich lab, we are on it
12:06:25 <jki> anything else on testing?
12:06:39 <jki> 5
12:06:41 <jki> 4
12:06:43 <jki> 3
12:06:45 <jki> 2
12:06:47 <jki> 1
12:06:49 <jki> #topic AOB
12:06:59 <jki> two topics from my side:
12:07:24 <jki> the extra budget for AI instances is approved now
12:07:46 <jki> so, we could start with re-deploying Uli's setup
12:08:52 <uli> i'm atm working on fixing it up so that it's a tiny bit more user-friendly
12:08:55 <uli> --help and such...
12:09:15 <jki> great! what support would you need after that?
12:09:33 <jki> or would anyone be able to set it up independently?
12:09:56 <uli> i guess the best would be to have a llama.cpp instance (or two) running somewhere that people can connect to.
12:10:03 <uli> and run the scripts on their local machine
12:11:01 <jki> and that llama.cpp would be what we host in AWS - can this be done in a on-demand way? Only spawning up on request?
12:11:31 <uli> no idea, to be honest. llama.cpp takes a moment to come up (many GBs of data to load), but it's not that bad.
12:11:52 <uli> from the sysadmin side i have no idea how stuff like that is done nowadays...
12:12:32 <jki> I can ask around for suggestions. Anyone with a proposal already, just speak up
12:13:28 <jki> anything else on this topic?
12:14:18 <jki> if not: the other thing is just the information that we will likely soon get access to OpenAI Codex Security
12:14:47 <jki> it may not be the best first step to try that on the CIP kernel, though, unless you already have some ideas
12:16:00 <jki> we will likely first try it on some userspace component that is critical as well
12:16:34 <jki> other topics for today?
12:17:07 <jki> 5
12:17:09 <jki> 4
12:17:11 <jki> 3
12:17:14 <jki> 2
12:17:16 <jki> 1
12:17:18 <jki> #endmeeting