13:00:59 <jki> #startmeeting CIP IRC weekly meeting 13:00:59 <collab-meetbot> Meeting started Thu Sep 4 13:00:59 2025 UTC and is due to finish in 60 minutes. The chair is jki. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:59 <collab-meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:59 <collab-meetbot> The meeting name has been set to 'cip_irc_weekly_meeting' 13:01:06 <jki> #topic AI review 13:01:14 <jki> currently none recorded 13:01:24 <jki> 5 13:01:26 <jki> 4 13:01:28 <jki> 3 13:01:29 <jki> 2 13:01:31 <jki> 1 13:01:33 <jki> #topic Kernel maintenance updates 13:01:46 <uli> i released 4.4 13:01:54 <masami> This week reported 3 new CVEs and 103 updated CVEs. 13:02:15 <pave1> I'm doing reviews, 6.1.44 and .45. 13:02:54 <iwamatsu_> hello 13:03:21 <jki> hi! we are at maintenance updates already 13:04:01 <iwamatsu_> I reviewed 6.12.42 and 43, 13:04:50 <arisut> currently kernel upstream is releasing new kernel as we talk 5.15.191,5.10.242, 5.4.298 13:04:58 <arisut> fyi 13:06:42 <jki> any further updates? 13:07:26 <jki> 5 13:07:28 <jki> 4 13:07:29 <jki> 3 13:07:31 <jki> 2 13:07:32 <jki> 1 13:07:36 <jki> #topic Kernel release status 13:07:51 <jki> 4.4-rt is late IIRC 13:08:29 <pave1> Yep, sorry about that. I don't foresee problems, should do it on friday or over weekend. 13:09:20 <jki> perfect 13:09:44 <jki> anything else on this? 13:09:55 <jki> 5 13:09:57 <jki> 4 13:09:59 <jki> 3 13:10:00 <jki> 2 13:10:02 <jki> 1 13:10:03 <jki> #topic Kernel testing 13:10:05 <arisut> o/ 13:10:26 <arisut> about release we have a way to mask problematic cip kernel ? 13:11:20 <pave1> Does that have anything to do with "currently upstream is releasing 5.4.298" announcement? :-) 13:11:32 <arisut> in the case that the kernel have build issues or similar 13:11:45 <arisut> just asking 13:12:18 <pave1> Not really, it may depend on severity. 13:12:37 <arisut> I see, thanks 13:14:20 <arisut> maybe could be useful to have something to indicate that a version have serious problems 13:14:20 <patersonc> On the testing side of things, we've got some really high load on the LAVA server which is causing some issues 13:15:03 <arisut> I talked with collabora about the issues and they are willing to share some filtering script 13:15:17 <arisut> problem is that we don't have root access to lava-cip servers 13:15:44 <patersonc> indeed 13:16:03 <arisut> :( 13:16:20 <arisut> another solution could be to ask LF to set cloudfare 13:16:30 <arisut> they already using it for kernel.org 13:16:58 <arisut> probably is the fastest solution as LF have already experience with that 13:17:08 <patersonc> Yea 13:17:20 <arisut> qq 13:17:28 <patersonc> I'll keep trying to narrow down the issue and talk to LF in tandem 13:17:32 <arisut> also broonie is using it for his lava server 13:17:57 <arisut> qq <- writing on the wrong keyboard lol 13:19:24 <jki> patersonc: no one reacted on your lava load issue yet, right? 13:19:45 <patersonc> Nope. ALthough Cybertrust lab admin said it wasn't them 13:19:55 <pave1> patersonc: https://gitlab.com/cip-project/cip-kernel/linux-cip/-/pipelines/2021753921 I'm having more problems than usual with testing. Is that related / likely to get better on its own? 13:20:13 <jki> we cannot identify the source?? 13:20:29 <jki> do all workers share the same token? 13:20:31 <arisut> collabora said is a know issue, there are some bots that are causing problems with lava 13:20:50 <jki> AI crawlers again, nothing authenticated? 13:21:15 <patersonc> It could just be web traffic, rather than api 13:21:43 <jki> can we close the web interface without affecting the API? 13:21:52 <arisut> without server access we can only suppose 13:21:53 <jki> at least temporarily 13:22:35 <jki> oh, that was the missing root access?! 13:22:46 <jki> who has then, LF admins? 13:22:48 <arisut> yes 13:22:59 <jki> -> LF ticket? 13:23:15 <patersonc> I've already got a discussion open with our usual LF IT contactr 13:23:21 <jki> ok 13:23:24 <arisut> patersonc, thanks ! 13:24:48 <jki> guess we have to wait and see then... 13:25:02 <jki> other testing topics? 13:25:16 <patersonc> For now, kernel testing may take a hit. There are still some labs offline since I rebooted the LAVA master 13:25:48 <pave1> Ok. I'll hit retry few more times ;). 13:25:48 <arisut> I'm starting to add nfsrootfs build and x86 platforms 13:26:36 <jki> only one is offline right now, iwamatsu's 13:27:16 <jki> but several in "maintenance" state - is that what you mean? 13:27:31 <patersonc> Yea 13:27:38 <iwamatsu_> maybe other lab does not reconnect yet. 13:27:41 <patersonc> I put them all into maintenance until the lab owners reboot the labs 13:28:07 <patersonc> LAVA doesn't like it when the server reboots. The labs don't automatically start working again - they want a reboot too 13:28:36 <jki> did you inform all lab owners already? 13:28:45 <patersonc> Yes 13:29:04 <arisut> patersonc, beaglebone could boot with nfsrootfs https://lava.ciplatform.org/scheduler/job/1305458 13:29:32 <arisut> but we have still problems with r8a7743-iwg20d-q7 https://lava.ciplatform.org/scheduler/job/1305459 13:29:46 <patersonc> Thanks arisut 13:30:18 <arisut> maybe we could add some boot commands? 13:30:56 <arisut> I see there is a extra_kernel_args: lsm=capability,landlock,yama,loadpin,safesetid,selinux,bpf on some nfs machine 13:31:23 <patersonc> I've never set those for the iwg20d boards before 13:31:31 <arisut> on the collabora boards 13:31:40 <arisut> maybe we need something similar 13:31:56 <arisut> I will investigate difference on the build 13:32:02 <patersonc> Thanks 13:32:25 <arisut> np 13:32:37 <iwamatsu> it can not boot with zImage 13:33:23 <patersonc> We need to work out why zimage isn't working 13:33:32 <iwamatsu> iwg20d need to use uImage with default bootloader 13:33:59 <patersonc> Arisut is there a way to get uimage out of kernelci? 13:34:01 <iwamatsu> meybe it depends on address of boot imge. 13:34:26 <arisut> patersonc, I doubt :/ will check 13:35:12 <patersonc> Thanks 13:35:37 <arisut> it cannot work with zimage or maybe it could work but is not tested ? 13:36:33 <patersonc> Maybe we just need to check the uboot command 13:36:39 <arisut> ok 13:37:21 <arisut> I will check about using uimage 13:37:44 <arisut> looks like we have a make target for uimage 13:37:54 <iwamatsu> I think the bootloader didn't support zImage... 13:38:18 <arisut> I hope that is still working on pipeline but looks like is not used anymore 13:38:58 <arisut> I will check how to make it throw uimage 13:39:05 <arisut> thanks 13:39:14 <patersonc> Thanks 13:39:21 <jki> patersonc: it's likely hopeless to wait for a denx lab reboot - just set it to active again and pray ;) 13:39:43 <patersonc> yea 13:40:01 <jki> Quirin set both our labs to active as well - works without rebooting 13:40:16 <jki> maybe that issue was with older lava 13:40:16 <pave1> jki: I can try to ask for reboot -- should I? 13:40:38 <jki> pavel: only if we see issues, I would say 13:40:41 <patersonc> jki: The issue is that the LAVA_DISPATCHER_IP variable doesn't get set. So NFS jobs will fail 13:40:43 <pave1> jki: ok. 13:43:37 <jki> patersonc: does not seem to be the case for our labs 13:46:03 <patersonc> okay 13:46:07 <patersonc> I'll keep an eye on it 13:46:30 <jki> yeah, just ping us 13:47:13 <jki> more testing topics? 13:47:32 <jki> 5 13:47:34 <jki> 4 13:47:36 <jki> 3 13:47:38 <jki> 2 13:47:40 <jki> 1 13:47:42 <jki> #topic AOB 13:48:48 <masami> I will be absent next week 13:49:17 <patersonc> Same here 13:50:41 <jki> ok 13:50:52 <jki> anything else for today? 13:51:09 <jki> 5 13:51:11 <jki> 4 13:51:13 <jki> 3 13:51:14 <jki> 2 13:51:16 <jki> 1 13:51:19 <jki> #endmeeting