IRC logs for #cip for Thursday, 2020-04-09

*** masashi910 has joined #cip04:04
*** rajm has joined #cip06:38
*** KaidenPK_Yu_MOXA has joined #cip07:35
*** KaidenPK_Yu_MOXA has left #cip07:36
*** tpollard has joined #cip08:02
*** KaidenPK_Yu_MOXA has joined #cip08:49
*** wens has joined #cip08:57
*** fujita3 has joined #cip08:59
*** suzuki12 has joined #cip08:59
masashi910#startmeeting CIP IRC weekly meeting09:00
brloggerMeeting started Thu Apr  9 09:00:00 2020 UTC and is due to finish in 60 minutes.  The chair is masashi910. Information about MeetBot at http://wiki.debian.org/MeetBot.09:00
brloggerUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.09:00
brloggerThe meeting name has been set to 'cip_irc_weekly_meeting'09:00
*** brlogger changes topic to " (Meeting topic: CIP IRC weekly meeting)"09:00
masashi910#topic rollcall09:00
*** brlogger changes topic to "rollcall (Meeting topic: CIP IRC weekly meeting)"09:00
masashi910please say hi if you're around09:00
szlinhi09:00
punithi09:00
KaidenPK_Yu_MOXAhi09:00
iwamatsuhi09:00
fujita3hi09:00
suzuki12hi09:00
wenshi09:00
patersonchi09:00
masashi910#topic AI review09:00
*** brlogger changes topic to "AI review (Meeting topic: CIP IRC weekly meeting)"09:00
masashi9101. Combine root filesystem with kselftest binary - Iwamatsu-san09:00
bwhhi09:01
iwamatsumasashi910: I do not have update to this.09:01
masashi910iwamatsu: Noted. Thanks.09:01
masashi9102. Strengthen sustainable process to backport patches from Mainline/LTS - Kernel Team09:01
masashi910The current discussion among the kernel team is described in the following document.09:01
masashi910https://docs.google.com/document/d/1U4-txCx_uofNH7YXc9LgKILdY-BPgb-nYm3DFrs-9Ns/edit?usp=sharing09:01
masashi910If you have any ideas, please write down to the document.09:02
masashi910so, let me move on.09:02
*** pave1 has joined #cip09:02
masashi9103. Upload a guideline for reference hardware platform addition - masashi91009:02
pave1hi09:02
masashi910Updates can be reported around June timeframe.09:02
masashi910so, let me move on.09:02
masashi9104. Identify target boards and configs for RT kernel testing (4.4rt and 4.19rt) - masashi91009:02
masashi910Last week, it was decided to identify target boards and configs for RT kernel testing.09:02
masashi910I sent out an inquiry mail to ask CIP members to respond by April 16 (Thu).09:02
masashi910So far only Renesas (Chris-san) responded. I will follow up at the TSC call next week.09:03
masashi910any other topics?09:03
masashi910309:03
masashi910209:03
masashi910109:03
masashi910#topic Kernel maintenance updates09:03
*** brlogger changes topic to "Kernel maintenance updates (Meeting topic: CIP IRC weekly meeting)"09:03
wensnothing special on my end this week09:04
pave1I did reviews on 4.19.114. Sorry for -rt delays.09:04
pave1I currently believe that adding patch similar to one we have in 4.19-rt to 4.4-rt is the best way forward, even if I can't test it properly.09:05
bwhI reviewed updates to kernel-sec and kernel-config09:05
iwamatsumasashi910: I reviewd 4.4.218 and I send to backport patches for 4.4.y to stable ML.09:05
patersoncThanks bwh:09:05
patersonc * Thanks bwh09:05
masashi910iwamatsu: thanks for adding (CIP) after your name when you sent to the stable ML!09:06
masashi910wens, pave1, bwh, iwamatsu: Thanks for your reports!09:07
masashi910any other topics?09:07
masashi910309:08
masashi910209:08
masashi910109:08
masashi910#topic Kernel testing09:08
*** brlogger changes topic to "Kernel testing (Meeting topic: CIP IRC weekly meeting)"09:08
masashi910patersonc: the floor is yours09:08
patersoncHello09:08
patersonc#info Added RT configs for RZ/G1 and RZ/G2 platforms to cip-kernel-config.09:08
patersonc#info New USB hubs arrived at Renesas yesterday. I plan to install them later today. Hopefully this will resolve the disconnection issues we currently have.09:08
patersonc#info lab-cip-denx is online now, although we're seeing an issue with corrupted u-boot commands on the beaglebone-black board.09:08
patersonc#info Started adding support for RT tests, starting with cyclictest, cyclicdeadline and hackbench.09:09
patersonc#info Main issue/annoyance at the moment is that the tests use a Python script to parse the results, which currently is missing from the RFS.09:09
patersonc#link https://gitlab.com/cip-project/cip-testing/linux-cip-ci/-/merge_requests/3609:09
patersoncIf there are any other RT tests that would be helpful please let me know.09:09
pave1Thanks a lot for the work on the RT tests.09:09
patersoncnp, sorry it took so long to get around to it ;)09:09
pave1I believe cyclictest+hackbench should be pretty good.09:10
patersoncJust let me know if there is anything you want tweaked09:10
patersonc(see my comments on GitLab)09:10
pave1If we can get numbers on de0-nano, I'll know more.09:10
pave1Yes, I need to take a deeper look.09:10
patersonchttps://lava.ciplatform.org/scheduler/job/1418109:11
masashi910patersonc: thanks for your works!09:11
punitpatersonc: There seems to be a driver for cyclictest - rteval (https://git.kernel.org/pub/scm/utils/rteval/rteval.git/). It supposedly generates background load09:11
patersoncBtw I'm off on "holiday" next week09:11
punitI wonder if it'll help with what you're doing09:12
szlinpatersonc: stay safe :D09:12
patersoncThanks punit, I'll take a look09:12
masashi910patersonc: so, please stay safe. do you have any other topics?09:13
patersoncpave1: cyclictest+hackbench on de0-nano-soc board: https://lava.ciplatform.org/scheduler/job/1427909:14
patersoncmasashi910: Nothing else from me09:14
masashi910patersonc: thanks again!09:14
masashi910are there any other queries?09:14
masashi910309:14
masashi910209:14
masashi910109:14
masashi910#topic CIP Core09:15
*** brlogger changes topic to "CIP Core (Meeting topic: CIP IRC weekly meeting)"09:15
masashi910punit: the floor is yours.09:15
punitThanks, masashi91009:15
punitOne of the action items from discussions around Embedded world was target EOL dates for CIP Core components.09:16
punitI sent an email to cip-dev to kickstart the discussion!09:16
punitOnce we have some agreement, I propose to update the wiki to capture this informatoin09:16
punitPlease let me know if there is any feedback on the topic09:17
punitAnother item -09:17
punitThere is a preference to use python3 to the extent possible.09:17
punitAs python2 is EOL it'll get hard to support it for long term09:18
punitThe discussion was driven by Security WG query regarding backup software09:19
punitThat's it from me. Thank you09:19
yoshidak[m]Thank you for concluding this, Punit!09:19
pave1If I may chime in.09:19
pave1It is hard to comment without knowing what exact requirements for backups are.09:20
pave1duplicity is complex piece of software IIRC, and it seems somehow interesting to be critical part of core system.09:20
pave1python2 to python3 port of large code base will be "interesting" and may need quite a bit of testing.09:20
punitI agree - yoshidak[m] had sent an excel sheet with requirements sometime back based on our request09:20
pave1(Or you may be lucky).09:21
punitBut it maybe worth extracting it from the excel to an easier to discuss format09:21
punitFor the second part, I hope yoshidak[m] can clarify09:22
punitbtw, upstream has migrated to python3 (also in bullseye)09:22
punitSo hopefully, the backport is easier. From what I could find, most of the dependencies (except librsync2) are already in buster09:23
punitBut agreed, somebody needs to take a deeper look at what porting actually involves09:23
punitAnd how to verify the port09:24
pave1...and if it is right for us to do the porting, as opposed to declaring "sorry but CIP can not serve that usecase".09:24
pave1...because CIP normally tries to package well-maintained software and maintain it for years...09:25
pave1...while this is doing cutting edge development.09:25
pave1I mean... it will be useful to have duplicity working with python3 as python2 will disappear one day,09:25
pave1but I'm not sure if it should be us doing so.09:26
punitGood point - it depends on how strong the requirement for duplicity (backup software) is in CIP09:26
punitAt the least, I would like to see it in buster-backports09:27
masashi910pave1, punit: thanks for the discussion. this is very important topic.09:27
masashi910and I think this kind of discussion is also initiated in the ML.09:28
masashi910we need to continue to discuss.09:28
yoshidak[m]I'll clarify the requirement about sytem-backup, and consider the necessity of duplicity as well.09:28
masashi910yoshidak[m]: thanks for your comment.09:29
masashi910if further discussion is needed, shall we do that after the irc or in the ML?09:29
punitMy suggestion would be to continue on mailing list to allow people time to research before replying09:30
masashi910punit: +109:31
masashi910punit: thanks for your report!09:31
masashi910shall we move on?09:31
masashi910309:31
masashi910209:31
masashi910109:32
masashi910#topic Software update09:32
*** brlogger changes topic to "Software update (Meeting topic: CIP IRC weekly meeting)"09:32
masashi910Suzuki-san, the floor is yours09:32
suzuki12Hello.09:32
suzuki12I'm afraid I don't have major updates since last meeting.09:32
suzuki12That's it from me.09:32
masashi910suzuki12: noted. thanks for your report.09:32
masashi910are there any queries?09:32
masashi910309:33
masashi910209:33
masashi910109:33
masashi910#topic CIP Security09:33
*** brlogger changes topic to "CIP Security (Meeting topic: CIP IRC weekly meeting)"09:33
masashi910Yoshida-san, the floor is yours09:33
yoshidak[m]We are waiting to get approved the budget to start the gap assessment with the certification body.09:33
yoshidak[m]Next governing board meeting will be held at 27 or 28 April, so we are confirming to Exida about the valid period of the estimate.09:33
yoshidak[m]In parallel, we started to prepare the documentation for certification.09:33
yoshidak[m]Documents created for certification will be the main outcomes from the Security WG for users.09:34
yoshidak[m]However, we have just begun to consider what is needed.09:34
yoshidak[m]Hope to start discussion w/ Exida soon.09:34
yoshidak[m]That's all from me in this week.09:34
masashi910yoshidak[m]: thanks for your report.09:35
masashi910are there any queries?09:35
masashi910309:35
masashi910209:35
masashi910109:35
masashi910#topic AOB09:35
*** brlogger changes topic to "AOB (Meeting topic: CIP IRC weekly meeting)"09:35
masashi910Are there any business matters to discuss?09:35
masashi910309:35
masashi910209:35
masashi910109:36
masashi910#endmeeting09:36
brloggerMeeting ended Thu Apr  9 09:36:04 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)09:36
brloggerMinutes:        https://irclogs.baserock.org/meetings/cip/2020/04/cip.2020-04-09-09.00.html09:36
brloggerMinutes (text): https://irclogs.baserock.org/meetings/cip/2020/04/cip.2020-04-09-09.00.txt09:36
brloggerLog:            https://irclogs.baserock.org/meetings/cip/2020/04/cip.2020-04-09-09.00.log.html09:36
*** brlogger changes topic to "Civil Infrastructure Platform Project. Find the logs at https://irclogs.baserock.org/cip/"09:36
masashi910Thank you! Stay safe!09:36
pave1Thank you!09:36
szlinthank you09:36
iwamatsuthank you!09:36
suzuki12Thank you!09:36
punitThank you!09:36
KaidenPK_Yu_MOXAThank you!09:36
fujita3Thank you!09:36
wensThank you!09:36
*** suzuki12 has quit IRC09:36
bwhthanks09:36
*** fujita3 has quit IRC09:38
pave1patersonc: Thanks for all the realtime work.09:39
patersoncno worries09:39
patersoncHopefully it'll help your work09:39
pave1patersonc: It seems good enough for a start.09:39
pave1patersonc: Can it be integrated in such a way that "maximum latency" over 100usec fails the test, else it passes?09:40
patersoncGood09:40
patersoncYes, but the current setup relies on python to process this09:40
pave1Will I need to modify .yml files in the -rt trees?09:40
patersoncNo need for .gitlab-ci.yaml updates. I can handle that in the linux-cip-pipelines repo, once the current MR is merged.09:41
pave1And problem is "there's no python in the images" or "we need python2 and only have python3"?09:41
pave1I'll need to take a look at cyclicdeadline test.09:43
patersoncI don't think there is any python in the current images, although I'm not sure I've looked/tried the isar images yet09:43
patersoncpave1: There are also other RT tests, take a look through https://github.com/Linaro/test-definitions/tree/master/automated/linux and take your pick09:44
pave1Thanks, will take a look.09:45
pave1BTW good luck with the virus thing. It looks like it will be very interesting in next week or so for you.09:45
pave1We managed to "pause" it for now, which is either very good or very bad thing, depending on what happens next, but it is currently boring here.09:46
patersoncYea. Just need to stay in doors and hope we don't get anything when we go shopping09:46
pave1Yes, shopping is a bit scary :-(. Outdoors away from other people should not actually be bad. (At least here, we still have < 0.1% infected).09:49
pave1I don't understand what cyclicdeadline is.. all I see is binaries.09:50
pave1Usual non-rt tests still make sense to run in -rt configs, but I guess we are doing that.09:51
pave1cyclictest+hackbench should be okay for now.09:51
pave1Eventually, I'd expect -cip-rt users to ask for particular tests...09:52
patersonccyclicdealine description, see: https://github.com/Linaro/test-definitions/blob/master/automated/linux/cyclicdeadline/cyclicdeadline.yaml#L409:52
pave1Aha, thanks!09:52
pave1Okay, so that one does make sense to run, and I assume it will also need hackbench in the background to be useful.09:53
patersoncProbably09:54
patersoncI'm going afk for a while. Can you make any requests on the MR?09:54
pave1Or actually.. this is basically scheduler test. So it we need cyclictest _or_ cyclicdeadline, but unless debugging scheduler09:54
pave1... this is not critical. And I don't plan to debug scheduler :-).09:54
pave1Ok, will do;09:54
pave1but the best way would be to just merge it :-).09:54
pave1See you!09:55
*** KaidenPK_Yu_MOXA has quit IRC10:15
*** pave1 has quit IRC10:35
*** rajm has quit IRC13:24
*** rajm has joined #cip14:28
*** tpollard has quit IRC16:58
*** pave1 has joined #cip19:07
*** pave1 has quit IRC19:36
*** rajm has quit IRC22:20

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!