12:02:43 <jki> #startmeeting CIP IRC weekly meeting 12:02:43 <collab-meetbot`> Meeting started Thu Nov 3 12:02:43 2022 UTC and is due to finish in 60 minutes. The chair is jki. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:02:43 <collab-meetbot`> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 12:02:43 <collab-meetbot`> The meeting name has been set to 'cip_irc_weekly_meeting' 12:02:49 <jki> #topic AI review 12:02:54 <jki> 1. Add qemu-riscv to cip-kernel-config - patersonc 12:03:06 <jki> no one here today to address this 12:03:15 <jki> other AIs I missed? 12:03:41 <jki> 3 12:03:43 <jki> 2 12:03:45 <jki> 1 12:03:48 <jki> #topic Kernel maintenance updates 12:03:56 <uli_> still reviewing 5.10.150 12:04:03 <masami> This week reported 6 new CVEs and 3 updated CVEs. 12:04:25 <iwamatsu> I reviewd 5.10.152 12:08:19 <jki> seems we don't have much to coordinate here this week 12:08:57 <jki> anything else on maintenance? 12:09:27 <jki> 3 12:09:28 <jki> 2 12:09:31 <jki> 1 12:09:34 <jki> #topic Kernel testing 12:10:01 <alicefm> No updates this week 12:10:17 <fbezdeka> As Pavel is not here today we might have to skip it: we still have https://github.com/kernelci/kernelci-core/issues/1053 open. My understanding was that the test report for the 4.4-cip tree/branch was usable now. I guess we can close it soon... 12:10:30 <pave1> I'm here now. Sorry. 12:10:33 <pave1> DST kicked in. 12:11:09 <jki> hi, pavel! 12:11:22 <pave1> Hi! 12:11:40 <jki> can you quickly catch up - anything to add on the previous topic or on the testing question? 12:12:24 <pave1> I was reviewing 5.10.152 and am reviewing 5.10.153. Otherwise not much updates from me. 12:12:45 <pave1> I need to take another look at 4.4 testing effort, but feel free to close github issue. 12:13:50 <pave1> gitlab does not work well today: https://gitlab.com/cip-project/cip-kernel/linux-cip/-/pipelines/684313254 12:14:15 <pave1> The tasks with red status have already had 3 tries... 12:15:14 <jki> ERROR: Job failed (system failure): pods "runner-ufmsyps-project-2678032-concurrent-02pt99" not found - Huh?! 12:15:55 <pave1> Yep, that happens a lot. I normally just hit retry :-). 12:16:03 <pave1> 4.4 testing -- there may be still some work to do on 4.4 12:16:08 <pave1> Latest reports I could find: 12:16:11 <pave1> cip/linux-4.4.y-cip smc: 12 runs, 3 regressions (v4.4.302-cip69-517-g92709d8ae585) #kerne 12:16:31 <pave1> cip/linux-4.4.y-cip baseline: 102 runs, 6 regressions (v4.4.302-cip69-517-g92709d8ae585 12:16:43 <pave1> cip/linux-4.4.y-cip ltp-pty: 3 runs, 1 regressions (v4.4.302-cip69-517-g92709d8ae585) #ke 12:16:55 <pave1> I know smc is problematic, but we have regressions elsewhere, too. 12:18:31 <pave1> One of failures is: https://storage.kernelci.org//cip/linux-4.4.y-cip/v4.4.302-cip69-517-g92709d8ae585/arm64/defconfig/gcc-10/lab-broonie/baseline-qemu_arm64-virt-gicv2.html 12:19:37 <pave1> I see no kernel logs, so I suspect it is something wrong with the test system. 12:19:44 <fbezdeka> At least some (if not all) of the smc failures are reported by qemu targets. I guess that's more a common kernelci testing issue 12:19:50 <pave1> So perhaps don't close the gitlab issue just yet :-). 12:24:32 <jki> how to proceed here, who can help / needs to help? 12:27:24 <iwamatsu> I re-run pavel's pipeline. Some jobs have worked and succeeded. 12:27:54 <fbezdeka> Build tests are looking good, only a few warnings. That might be something for Pavel to look at. I could try to talk to kernelci if testing smc on virtual targets has any value 12:28:14 <iwamatsu> However, the following jobs have a problem with CI. 12:28:22 <iwamatsu> https://gitlab.com/cip-project/cip-kernel/linux-cip/-/jobs/3268935753 12:28:23 <pave1> Yep, disabling smc on qemu might be good idea. 12:28:33 <pave1> What about baseline and ltp-pty regressions? 12:29:03 <fbezdeka> That would be the next steps. 12:30:18 <iwamatsu> about qemu/smc issue, it is not difficult. I will create MR for that. 12:30:21 <pave1> fbezdeka: And for the record, I know about the ~ 1 warning,just was busy with other stuff. Sorry about that. 12:31:03 <fbezdeka> It's just a warning. I'm quite sure it doesn't bring up something serious 12:31:57 <fbezdeka> qemu/smc: Maybe someone else should confirm that it has no value to test microcode related stuff on virtual targets. Maybe I/we miss something 12:32:35 <pave1> fbezdeka: IIRC... trouble is that we are using KVM or something, so whether test works or not depends on configuration of host. 12:33:10 <fbezdeka> Hm... But even with KVM enabled we would get random results (depends on the job scheduling I guess) 12:33:27 <pave1> fbezdeka: I suggest disabling smc on qemu. We have enough other stuff to work on. 12:33:42 <fbezdeka> Fine with that. 12:34:27 <iwamatsu> And qemu machine option. user can contorol it with qemu option. 12:35:56 <iwamatsu> ssbd spec-ctl etc.. 12:37:12 <iwamatsu> I use them to avoid the qemu/smc issue, but I'm still on the way. 12:39:36 <jki> ok, seems there is a plan how to proceed 12:39:45 <jki> anything else on this? 12:40:56 <jki> 3 12:40:58 <jki> 2 12:41:01 <jki> 1 12:41:05 <jki> #topic AOB 12:43:07 <jki> any topic? 12:43:39 <jki> 3 12:43:41 <jki> 2 12:43:43 <jki> 1 12:43:46 <jki> #endmeeting