09:00:01 <masashi910> #startmeeting CIP IRC weekly meeting 09:00:01 <brlogger> Meeting started Thu Mar 11 09:00:01 2021 UTC and is due to finish in 60 minutes. The chair is masashi910. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:00:01 <brlogger> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:00:01 <brlogger> The meeting name has been set to 'cip_irc_weekly_meeting' 09:00:04 <masashi910> #topic rollcall 09:00:10 <masashi910> please say hi if you're around 09:00:12 <pave1> hi 09:01:01 <masashi910> Today, Iwamatsu-san is not here. Let's get started. 09:01:02 <patersonc> hi 09:01:07 <masashi910> #topic AI review 09:01:15 <masashi910> 1. Combine root filesystem with kselftest binary - iwamatsu 09:01:22 <masashi910> According to Iwamatsu-san, "No update". 09:01:28 <masashi910> 2. Do some experiment to lower burdens on CI - patersonc 09:01:31 <patersonc> No updates 09:01:44 <masashi910> patersonc: Noted. Thanks. 09:01:55 <masashi910> Let's move on. 09:01:56 <masashi910> #topic Kernel maintenance updates 09:02:02 <masashi910> According to Iwamatsu-san, "I reviewed 4.4.260, 5.10.21 and 5.10.22". 09:02:19 <pave1> I did reviews on 5.10.21, 22, 23 and related 4.19 kernels". 09:02:38 <pave1> Oh and the camera series from the Renesas. 09:02:49 <masashi910> pave1: Thanks for your works! 09:02:58 <masashi910> wens: are you around? 09:03:01 <wens> hi 09:03:09 <masashi910> wens: Hi! 09:03:12 <wens> I'm still working on the report, sorry. 09:03:25 <wens> Summary: seven new issues this week, all fixed. 09:03:36 <pave1> wens: Boring report. Good! :-) 09:03:41 <wens> :) 09:04:10 <masashi910> wens: Thanks for your works! 09:04:21 <masashi910> any other topics? 09:04:33 <masashi910> 3 09:04:38 <masashi910> 2 09:04:41 <masashi910> 1 09:04:44 <masashi910> #topic Kernel testing 09:05:04 <patersonc> The main issue is that lab-cip-renesas is down, which means that the CI is failing all over the place due to lack of Renesas boards 09:05:05 <patersonc> Sorry about this 09:05:34 <patersonc> We use a seperate network for the lab and a big power outage did funny things to the old router we're using 09:05:39 <patersonc> A new one is on order 09:05:46 <wens> it happens :( 09:06:09 <patersonc> Ironically this is exactly why we're trying to get boards into multiple labs 09:06:38 <patersonc> We have some that have arrived at Denx, but covid has slowed down getting them installed 09:07:03 <patersonc> Sorry if this gets in the way of the planned releases tomorrow 09:07:25 <patersonc> If needed I can run some manual testing in our internal lava lab on the Renesas boards? 09:07:43 <pave1> Are you talking to Denx? Maybe the Covid situation got better there. 09:07:58 <patersonc> pave1: I'll send an email now and check 09:08:37 <pave1> It would be iwamatsu doing the releases. But yes, I guess testing them would be good idea. 09:09:13 <masashi910> patersonc: Thanks for your works. 09:09:26 <masashi910> patersonc: I have a question. 09:09:38 <masashi910> Cybertrust is planning to propose Xilinx zcu102 as an official platform. 09:09:46 <masashi910> According to my understanding, currently, as a candidate platform, boot testing and Spectre/Meltdown checking are executed on the boards. 09:09:56 <masashi910> Once it is accepted as an official, LTP will be executed on the boards as well? 09:10:38 <patersonc> Ideally we'd test LTP on all boards. It's just a question of making a suitable filesystem and updating a script or two 09:11:01 <patersonc> And if the LTP tests take a while to run - making sure we have enough boards in the labs 09:11:44 <pave1> Doing LTP run manually on that board might be good idea. 09:11:51 <pave1> But I don't expect problems there. 09:12:06 <masashi910> As you know, currently CTJ lab has two zcu102. Our target branch is 5.10 only. 09:12:39 <patersonc> masashi910: Although currently there is only 1 online? https://lava.ciplatform.org/scheduler/device_type/zynqmp-zcu102 09:13:02 <patersonc> masashi910: Would you be able to send 2 boards to Denx to put in their lab? 09:13:42 <masashi910> patersonc: zcu102 is costly. I am afraid we cannot afford to do so. 09:14:22 <masashi910> patersonc: I will check the status about the other one internally. 09:14:43 <patersonc> Ah yes I remember now. That's okay 09:15:00 <masashi910> patersonc: Thanks. 09:15:09 <masashi910> any other topics? 09:15:18 <masashi910> 3 09:15:21 <masashi910> 2 09:15:25 <masashi910> 1 09:15:27 <masashi910> #topic CIP Security 09:15:37 <masashi910> Yoshida-san, are you around? 09:16:18 <masashi910> So, let's skip. 09:16:20 <masashi910> #topic AOB 09:16:26 <masashi910> Are there any business to discuss? 09:16:37 <pave1> I guess we should talk about the zcu102. 09:16:50 <masashi910> pave1: Sure. 09:17:05 <wens> No progress made on the Gitlab cip-kernel-sec merge request rebase by me yet. 09:17:25 <pave1> The plan with 5.10 seems sane, given that the board already works in 5.10. 09:17:32 <masashi910> wens: Sure. Thanks. 09:17:58 <pave1> Might be nice to call it zcu102 in all the relevant places, people may not realize zynqmp refers to same board. 09:18:12 <pave1> I believe we should just go ahead. 09:18:51 <masashi910> pave1: Thanks for your comments. If things go well, I can propose this at the next TSC, next Tuesday. 09:19:00 <wens> Another thing, futex were broken in v4.9.y, and there were some additional patches that were backported to v4.4.y, is anyone tracking those? 09:19:01 <pave1> You have my support for that. 09:19:30 <masashi910> pave1: And yes, we should change the name approriately. 09:19:41 <masashi910> pave1: Thanks very much for your support!!! 09:21:19 <pave1> wens: No idea about 4.4. If you have pointers, I can take a look... or maybe we can do it over the email. 09:22:20 <wens> pave1: latest ones I've seen: https://lore.kernel.org/stable/20210309030605.3295183-1-zhengyejian1@huawei.com/ 09:23:23 <wens> v2 here: https://lore.kernel.org/stable/20210311032600.2326035-1-zhengyejian1@huawei.com/ 09:23:49 <pave1> Thanks! 09:24:02 <pave1> I guess... let me take a look and discuss over the mailing list. 09:24:13 <wens> thanks 09:24:15 <masashi910> wens: Thanks for raising this issue. 09:24:27 <masashi910> wens, pave1: Should I make this as AI? 09:24:34 <wens> there's no definitive list of futex-related patches 09:24:53 <pave1> Huawei/upstream is taking look, I don't think we need an AI :-). 09:24:56 <wens> masashi910: I think they'll get queued up as normal 09:25:19 <masashi910> pave1, wens: Sure, thanks. 09:25:27 <masashi910> any other topics? 09:25:41 <wens> Here is today's cip-kernel-sec report: https://lore.kernel.org/cip-dev/CAGb2v66-bSg43E7k+HYJv0CniGDwrW3GP4ewysjr-zX1y2fo1g@mail.gmail.com/ 09:26:04 <masashi910> wens: Thanks very much as always. 09:26:31 <masashi910> Maybe, it's time to close the meeting. 09:26:36 <masashi910> #endmeeting