13:02:10 #startmeeting CIP IRC weekly meeting 13:02:10 Meeting started Thu Sep 9 13:02:10 2021 UTC and is due to finish in 60 minutes. The chair is jki. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:02:10 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:02:10 The meeting name has been set to 'cip_irc_weekly_meeting' 13:02:22 #topic rollcall 13:02:26 hi 13:02:27 hi everyone 13:02:30 hi 13:02:31 o/ 13:02:33 hi 13:02:35 Hi 13:03:00 great, almost full-house 13:03:11 is chris going to join? 13:03:16 hi 13:03:33 perfect 13:03:48 anyone missing? 13:04:09 iwamatsu 13:05:44 well, let's start nevertheless I would say 13:05:50 Yep :-). 13:05:51 #topic AI review 13:06:08 1. Combine root filesystem with kselftest binary - iwamatsu & alicef 13:06:13 any news? 13:06:31 I don't think so 13:06:38 Investigating on how to expand the cip-core for make kselftest work on the user space 13:07:15 expand cip-core space 13:07:20 ok - i know it's being used already, I suppose you saw the recipe 13:07:30 yes 13:07:47 but I don't have details either -> maybe contact our mentor colleagues 13:08:02 e.g. via isar ML 13:08:13 yes 13:08:44 good 13:08:58 any new AIs? 13:09:52 I think I forgot one from last week: 13:09:58 Add lava URLs to wiki page 13:10:47 patersonc: think you wanted to check that, right? 13:12:14 I can't remember the details for this 13:13:02 kernelci.ciplatform.org and cip.kernelci.org: add to wiki once ready 13:13:26 jki: cip.kernelci.org is already ready 13:13:43 Ah okay 13:13:44 Thanks 13:13:44 I'll get that done 13:14:00 TIA! 13:14:07 other AIs? 13:14:19 3 13:14:23 2 13:14:28 1 13:14:41 #topic Kernel maintenance updates 13:15:07 I have reviewed patches queued for 5.10.62/63/64. 13:15:19 fixed a couple of minor issues in the scripts in cip-kernel-config and lts-commit-list 13:15:32 There is 3 new CVEs in this week. 13:15:40 CVE-2021-3715, CVE-2021-3759, CVE-2021-40490 are new CVEs. 13:15:45 These CVEs have been fixed in the mainline and some stable kernels. 13:16:52 CVE-2021-3739, CVE-2021-3743, CVE-2021-3753, CVE-2021-38198, CVE-2021-3444, CVE-2021-3600 have updated. 13:16:57 They have been fixed in the mainline and some stable kernels. 13:18:11 also for all our CIP kernel versions? 13:18:33 or how do those report map on our kernels generally? 13:19:15 jki: not all cip kernels are fixed. 13:19:42 jki: We don't have to do anything this week. 13:19:47 ok 13:20:03 jki: CVE-2021-3759: is missing cgroup accounting -- DoS only if you have shell. 13:20:04 still learning the meanings ;) 13:20:17 jik: This report is based on mainline and stable kernels. 13:21:02 CVE-2021-38198: is KVM nastyness. It is unclear if we want to track this one. Anyway someone backported it to 4.14, so things are improving. 13:21:59 ok - move on? 13:22:20 3 13:22:21 2 13:22:22 1 13:22:23 :-) 13:22:25 #topic Kernel testing 13:23:06 cip KernelCI production can now be seen at https://cip.kernelci.org 13:23:14 We've been working on various topics for the KernelCI kackfest 13:23:15 hackfest 13:23:18 smc (spectre meltdown check) 13:23:20 got merged and is now working on production 13:23:23 example: https://linux.kernelci.org/test/plan/id/6138e379ab9d472d1ad596a8/ 13:23:37 currently some cip configuration are used on staging kernelci (currently only for qemu arm and x86-64) 13:23:38 configuration are from https://gitlab.com/cip-project/cip-kernel/cip-kernel-config/ 13:24:27 patersonc[m] and I are currently fixing the kernelci pages links about and reference about CIP 13:25:22 like https://static.staging.kernelci.org/docs/org/tsc/#cip-instance and https://static.staging.kernelci.org/docs/instances/ 13:25:40 nice 13:26:03 currently I'm working with the help of patersonc[m] on making kernelci work with the isar-cip-core images 13:26:38 also I'm adding iec-security-tests as kernelci test 13:26:38 ok - i think / hope the preconditions are fulfilled 13:26:59 like being able to generate to right images (tarballs etc.) 13:27:13 that is what I'm working on yes 13:27:49 if something should be missing, we have other isar layers generating for lava, e.g. https://gitlab.denx.de/Xenomai/xenomai-images 13:28:08 mmm ? 13:28:36 if something is missing we cannot just add it to the isar-cip-core ? 13:28:51 yes, sure 13:29:08 that link was just another source of inspiration 13:29:13 I didn't know about xenomai-core 13:29:32 or ask on the list, we could help so that you don't need to learn isar first 13:29:39 and isar-cip-core 13:29:52 currently I'm making the script for deploying the images to kernelci fileserver 13:29:59 using the kernelci api 13:30:18 ok, that is surely specific 13:30:21 jki: thanks that would be helpful 13:31:01 drop an email with question, my colleague Quirin can likely help as well (he did that for xenomai) 13:31:29 creating tarball ? 13:31:48 including that, yes 13:32:08 yes ok 13:32:39 kas[-container] build ...:kas/opt/targz-img.yml 13:32:57 there are no other updates for me 13:33:06 yes 13:33:21 ok - anything else for this topic? 13:33:26 on isar just need to be enabled from .gitlab-ci 13:33:35 not at the moment 13:33:49 #topic AOB 13:34:03 one thing 13:34:16 i'm a "developer" member of all projects on gitlab now, but the master branches of cip-kernel-config and lts-commit-list are set to "protected" 13:34:25 apparently that means i need extra-special permissions to push there 13:34:32 who do i bother about that? 13:34:48 You definitely need permissons for lts-commit-list. 13:35:47 handling that 13:35:56 ok, thanks 13:36:03 how is cip-kernel-config maintained so far? 13:36:12 multi-maintainer? MRs? 13:36:15 iwamatsu is marked a 13:36:18 as a maintainer. 13:36:31 so he will merge to master? 13:36:58 I believe I have committed directly to it, but he's the one responsible... 13:37:54 then let's clarify this with him 13:38:03 maybe write him an email, uli? 13:38:09 can do 13:38:43 I believe Ulrich should get access there (same as lts-commit-list), but in this case it might be polite to ask before commiting. 13:39:01 right 13:39:12 lts-commit-list already done now 13:39:19 thx 13:39:30 good, any other access issues to clarify? 13:39:37 not from me 13:40:04 or other AOBs? 13:40:07 We should probably create shared whiteboard somewhere to coordinate patch reviews. 13:40:31 Not sure if git is suitable for that... 13:40:48 Ideally something that can be easily scripted... 13:41:09 If someone has quick idea, speak now, otherwise I'll write email to the list. 13:41:14 would issue tracking work (with email feeding)? 13:41:31 I think I'd avoid email for this. 13:42:06 I was thinking more like wiki page. We don't really need history, we just need shared place where people can write to easily. 13:42:44 wiki is generally with history... 13:43:07 jki: I know. I guess we can live with history but don't need it. 13:43:33 ...and is actually in git, if using one from gitlab.com 13:43:48 but maybe start with describing the workflows you have in mind 13:43:59 Ok, lets do that over email? 13:44:04 fine for me 13:44:49 anything else for today? 13:45:24 3 13:45:27 2 13:45:30 1 13:45:41 thank you all, have a nice day! 13:45:47 #endmeeting