09:00:01 <masashi910> #startmeeting CIP IRC weekly meeting 09:00:01 <brlogger`> Meeting started Thu Apr 1 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' 09:00:05 <masashi910> #topic rollcall 09:00:11 <masashi910> please say hi if you're around 09:00:14 <patersonc> hi 09:00:16 <wens> hi 09:00:19 <pave1> hi 09:00:46 <iwamatsu> hi 09:01:09 <masashi910> Let's get started. 09:01:17 <masashi910> #topic AI review 09:01:25 <masashi910> 1. Combine root filesystem with kselftest binary - iwamatsu 09:01:28 <iwamatsu> no update 09:01:42 <masashi910> iwamatsu: Noted, thanks! 09:01:49 <masashi910> 2. Do some experiment to lower burdens on CI - patersonc 09:02:00 <patersonc> no updates 09:02:14 <masashi910> patersonc: Noted, thanks! 09:02:23 <masashi910> 3. Ask board owners(siemens_i386-rt/plathome_obsvx1) whether X.25 and floppy can be disabled - masashi910 09:02:29 <masashi910> I got the confirmations from Minda-san and Jan-san respectively, so we can disable X.25 and floppy. 09:02:36 <masashi910> iwamatsu: Could you disable them? 09:02:55 <iwamatsu> Sure, I will do it after this meeting. 09:03:25 <masashi910> iwamatsu: Thanks! Then, I will close this AI. 09:03:37 <masashi910> 4. Monitor the status of CVE-2021-3444 and CVE-2021-20292 - Kernel Team 09:03:42 <masashi910> According to Wens-san's report: 09:03:55 <masashi910> ==== 09:03:55 <masashi910> CVE-2021-3444 - Debian added the following notes: 09:03:56 <masashi910> This last pre-requisite commit though would depend on 092ed0968bb6 ("bpf: verifier support JMP32") 09:03:56 <masashi910> which does not seem to make it possible to backport the fixes in 4.19.y easily. 09:04:04 <masashi910> CVE-2021-20292 - Ubuntu tagged the commit introducing the issue as 8e7e70522d76 09:04:04 <masashi910> ("drm/ttm: isolate dma data from ttm_tt V4") from v3.3-rc1. 09:04:04 <masashi910> So it looks like the fix needs to be backport to v4.4 as well. 09:04:04 <masashi910> ==== 09:04:22 <masashi910> So, shall we continue to monitor them? 09:04:36 <wens> I think so. 09:05:38 <masashi910> wens: Sure. So I will keep this AI open. 09:05:47 <masashi910> any other topics? 09:05:55 <masashi910> 3 09:05:59 <masashi910> 2 09:06:03 <masashi910> 1 09:06:06 <masashi910> #topic Kernel maintenance updates 09:06:20 <pave1> I have reviewed patches queued for 5.10.27. 09:06:29 <wens> This week' s report: https://lore.kernel.org/cip-dev/CAGb2v65SkZpyGfFGF51NSDjWj=rH1m8eddYy7gy4y+RA5j8nWQ@mail.gmail.com/ 09:07:12 <wens> 9 new issues this week, all fixed. Two need backporting, but one can be ignored for CIP. Remaining one needing backport is CVE-2021-29650. 09:07:36 <iwamatsu> I reviewed 4.4.254 and v5.10.27. 09:08:32 <masashi910> pave1, wens, iwamatsu: Thanks for your works! 09:08:57 <masashi910> Shall I make an AI to monitor CVE-2021-29650 status, or does anyone pick it up to work on? 09:10:28 <masashi910> So, I will make an AI for this CVE for now. 09:10:47 <masashi910> If it is not needed, please let me know. 09:10:54 <masashi910> any other topics? 09:11:08 <masashi910> 3 09:11:11 <masashi910> 2 09:11:14 <masashi910> 1 09:11:17 <masashi910> #topic Kernel testing 09:11:25 <masashi910> patersonc: The floor is yours. 09:11:41 <patersonc> Today the LAVA master will be offline for an hour or so due to maintenance 09:11:46 <patersonc> In about 5 hours 09:12:05 <patersonc> That said, some of the labs are offline this morning - they all go disconnected last night 09:12:08 <patersonc> *got 09:12:23 <patersonc> Sorry for the downtime 09:13:06 <patersonc> That's it from me 09:13:15 <masashi910> patersonc: Thanks for your works! 09:13:26 <masashi910> Any queries or comments? 09:13:35 <masashi910> 3 09:13:39 <masashi910> 2 09:13:43 <masashi910> 1 09:13:46 <masashi910> #topic CIP Security 09:13:54 <masashi910> yoshidak[m]: Are you around? 09:15:01 <masashi910> Because Yoshida-san does not seem to be here, let's skip. 09:15:10 <masashi910> #topic AOB 09:15:16 <masashi910> I have two topics. 09:15:31 <masashi910> 1. Announcement of new reference platform - Xilinx zcu102 09:16:09 <masashi910> At the latest TSC, Xilinx zcu102 was approved to add to the reference platform list. 09:16:37 <masashi910> Now it is going to be supported for 5.10/5.10-rt. 09:17:12 <masashi910> 2. Target boards for 5.10 and 5.10-rt 09:17:25 <masashi910> The 5.10 target boards were agreed as follows: 09:17:32 <masashi910> BBB, Cyclone V, QEMU(X86_64, Armv7, Armv8), RZ/G1M, RZ/G2M, SIMATIC IPC227E, OpenBlocks IoT VX2, Xilinx ZCU102 09:17:41 <masashi910> Regading 5.10-rt, SIMATIC IPC227E and Xilinx MPSoC ZCU102 are requested by CIP-members. 09:17:50 <masashi910> The kernel Team would like to add QEMU for testing. 09:18:05 <masashi910> Chris-san suggests to add RZ/G1M and RZ/G2M to the 5.10-rt targets. 09:18:17 <masashi910> patersonc: So, I will report that RZ/G1M and RZ/G2M are added to 5.10-rt targets at the next TSC. Is it OK? 09:18:46 <patersonc> That's okay with me 09:18:47 <patersonc> Thanks 09:18:57 <masashi910> patersonc: Sure. 09:19:06 <masashi910> So, any comments on the above? 09:19:25 <pave1> Cyclone V board is somehow imporant for me, as I have physical access to that one. 09:20:02 <masashi910> pave1: for 5.10-rt? 09:20:07 <pave1> But I'd propose simply adding all available boards for 5.10-rt testing (even if we don't support them, and even if we don't have realtime config for them). 09:20:32 <pave1> That would make it more likely that mistakes are caught, and should not really add much overhead. 09:20:39 <pave1> Yes, for 5.10-rt. 09:21:16 <patersonc> There are some non-CIP reference platforms in the Renesas lab, but only 1 of each. The issue comes in that pipelines will start failing if these boards go offline 09:21:59 <iwamatsu> +1 > all available boards. 09:22:26 <pave1> patersonc: So that's quite easy to diagnose -- if the results say "no available boards", it is not a "real" failure and can be ignored. 09:22:41 <patersonc> Okay :) 09:22:50 <pave1> patersonc: Yes, it would be nice to have different symbol than fail eventually, but... 09:23:19 <patersonc> Yes, GitLab doesn't support it though. If when CIP get a KernelCI instance things like this will be much better 09:23:22 <patersonc> I just haven't had time, sorry 09:23:53 <masashi910> pave1, patersonc, iwamatsu: I would like to confirm that this discussion is for testing, not official targets for 5.10-rt. 09:23:57 <pave1> No problem :-). 09:24:14 <pave1> masashi910: Yes, testing only. 09:24:27 <masashi910> pave1: Thanks for your confirmation! 09:25:03 <masashi910> So, the conclusion is to add all boards for 5.10-rt. Is it OK? 09:25:28 <pave1> I believe so. 09:25:42 <iwamatsu> It may be good to separate official and non-official. 09:26:23 <wens> as far as CVEs are concerned, I only look at cip-kernel-config 09:27:03 <masashi910> iwamatsu: That's correct. Officials are SIMATIC IPC227E and Xilinx MPSoC ZCU102. Plus, RZ/G1M and RZ/G2M. Non-officials are the rest. 09:27:16 <pave1> wens: Configs for -rt should be similar enough that it should not cause problems. 09:27:57 <wens> ok. If separate defconfigs get added, maybe we tag them as "test-only" then 09:28:34 <masashi910> Thanks for these discussions! Any other topics? 09:29:18 <masashi910> 5 09:29:22 <masashi910> 4 09:29:25 <masashi910> 3 09:29:28 <masashi910> 2 09:29:31 <masashi910> 1 09:29:33 <masashi910> So, let's close today's meeting. 09:29:38 <masashi910> #endmeeting