15:00:17 <pwhalen> #startmeeting Fedora ARM and AArch64 Status Meeting
15:00:17 <zodbot> Meeting started Tue Aug 17 15:00:17 2021 UTC.
15:00:17 <zodbot> This meeting is logged and archived in a public location.
15:00:17 <zodbot> The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:17 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:17 <zodbot> The meeting name has been set to 'fedora_arm_and_aarch64_status_meeting'
15:00:18 <pwhalen> #chair pwhalen pbrobinson coremodule
15:00:18 <zodbot> Current chairs: coremodule pbrobinson pwhalen
15:00:18 <pwhalen> #topic roll call
15:00:18 <pwhalen> good morning folks, who's here today?
15:00:28 * pbrobinson o/
15:00:31 <coremodule> hello!
15:00:34 <pwhalen> \o
15:00:42 <pwhalen> howdy coremodule
15:02:57 <jlinton> \me waves
15:03:51 <pwhalen> The usual suspects are all here..
15:03:57 <pwhalen> #topic 1) ==== Userspace Status  ====
15:04:21 <pwhalen> Any new user space issues?
15:05:38 <pwhalen> I have a few bugs filed on the Jetson Nano display issues, new one filed this morning
15:06:06 * nirik waves
15:06:27 <pwhalen> #info [Bug 1994582] [abrt] xorg-x11-server-Xwayland: strlcpy(): Xwayland killed by SIGABRT
15:07:44 <pwhalen> Has anyone hit any issues?
15:08:00 <pbrobinson> not that I've seen
15:08:12 <pbrobinson> that looks like it wouldn't be device specific
15:08:42 <pwhalen> Maybe not, will check some others today
15:08:55 <pwhalen> Nano was the first victim
15:09:37 <pwhalen> Ok, be great to get more eyes on things, all testing is appreciated
15:10:03 <pwhalen> #topic 2) ==== Kernel Status ====
15:10:13 <pwhalen> #info kernel-5.14.0-0.rc6.46.fc35
15:10:13 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1819587
15:10:47 <pwhalen> armhfp build looks like its hitting the kojid issue
15:11:47 <pwhalen> Any issues with 5.14 RC5 testing?
15:12:56 <pbrobinson> not with the little bit I've done
15:13:00 <jlinton> There is a CPPC FEI bugfix, that should be merged soon.
15:13:28 <pwhalen> #info Please test and report any issues to the list or #fedora-arm.
15:13:38 <jlinton> It should be visible on any arm64/ACPI machine with CPPC
15:13:51 <pbrobinson> jlinton: got a link to the upstream thread?
15:13:57 <jlinton> Yah, let me find it.
15:14:12 * nirik is open to any ideas on the lpae koji builders oom fun.
15:14:41 <jlinton> https://lore.kernel.org/lkml/20210804135925.3734605-1-dietmar.eggemann@arm.com/
15:15:58 <pbrobinson> nirik: yes, I have it on my list to look at that and reach out to arm people for assistance
15:15:59 <pwhalen> nirik: with the return of the kojid/builder issue, are we able to get them back to the combo that was working?
15:16:32 <nirik> pwhalen: the 'working' was no lpae kernel. Apparently when you dnf update on them it installs and defaults to the non lpae kernel
15:16:40 <pbrobinson> what's the difference between the last combo that was working, I think we had moved to a newer config that was working
15:16:42 <nirik> so, many builds were ok with 3GB
15:16:56 <nirik> but many were also failing that way (llvm in particular)
15:17:19 <nirik> pbrobinson: I think it was never working, just mistakenly not using lpae.
15:17:40 <nirik> which I could go back to, but that breaks llvm... but less than is currently broken I guess.
15:18:11 <nirik> might be an older kernel was better... not sure.
15:18:56 <pbrobinson> what do you mean by 3GB?
15:18:58 <pwhalen> I really thought we had a working combo, with lpae. It was only recently things started to break again.
15:19:13 <pbrobinson> lpae definitely used to work
15:19:40 <nirik> Well, I thought so too... but python was still affected...
15:20:22 <nirik> I can check kernel versions and times against reports from people
15:20:59 <nirik> it's really anoying tho. since I have no way to know until people start hammering on it, and many/most/some of them never report anything, their builds just restart a bunch of times and take longer
15:22:11 <pwhalen> nirik: right, *sigh*
15:22:48 <nirik> kernel-lpae-core-5.13.6-200.fc34.armv7hl      Tue 03 Aug 2021 04:04:35 PM UTC
15:22:48 <nirik> kernel-lpae-core-5.12.14-300.fc34.armv7hl     Fri 09 Jul 2021 06:27:03 AM UTC
15:22:58 <nirik> so might be worth going back to 5.12.14
15:23:10 <pwhalen> Definitely couldnt hurt
15:23:41 <pbrobinson> nirik: yes, I saw that with the u-boot build last night/this morning, although I pushed it late and didn't expect it to complete until long after I was snoring
15:23:42 <nirik> worth a shot. will try that and ask people to watch
15:24:04 <pbrobinson> nirik++ awesome, thanks
15:24:06 <zodbot> pbrobinson: Karma for kevin changed to 46 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
15:24:53 <pwhalen> thanks nirik
15:25:24 <pwhalen> #topic 3) ==== Bootloader Status ====
15:25:33 <pwhalen> #info uboot-tools-2021.10-0.2.rc2.fc36
15:25:33 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1819550
15:25:42 <pwhalen> New uboot, hot off the presses
15:25:43 <pbrobinson> The new RC2 I think fixes things
15:25:55 <pwhalen> Nice, will test today
15:26:06 <pbrobinson> there's been a bunch of migrations for the U-Boot DM stuff so please widely test devices to ensure no regression from upstream
15:26:30 <pwhalen> Ok, will do. Again, help appreciated.
15:27:08 <pbrobinson> I started looking at some of the beagles, there was a regression but will retest with the rc2 when I get a moment
15:28:36 <pwhalen> Thanks, I'll test as well.
15:28:40 <pwhalen> #topic 4) ==== Fedora 35 ====
15:28:51 <pwhalen> #info Latest nominated compose
15:28:51 <pwhalen> #link https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
15:29:21 <pbrobinson> IoT will be branched etc for f35 later today as an arm FYI
15:29:36 * pwhalen was going to ask :)
15:30:25 <pwhalen> This compose looked mostly ok, but still using rawhide repos, which has since been fixed
15:31:12 <pbrobinson> yea, I normally leave it a few days to settle down
15:32:56 <pwhalen> Any other issues in F35?
15:34:17 <pbrobinson> none off hand
15:34:26 <pwhalen> #topic 5)  == Open Floor ==
15:34:58 <nirik> I had a quick thing to bring up...
15:36:38 <pwhalen> nirik: all yours
15:36:40 <nirik> So, in the devel list thread where people were mad about that oom / kojid/ lpae thing it was brought up that perhaps we shoud look at dropping 32bit arm. I get the sense personally that it's too soon, but do we have any idea how popular 32bit arm stuff is now? is iot using it for anything? or is there other big uses that should be known? Or has everything moved pretty solidly to 64 bit now?
15:38:42 <pwhalen> nirik: I've had some discussions about planning to deprecate 32-bit, largely coming from the lack of hardware on the horizon.
15:38:48 <pbrobinson> yes, we have an IoT image
15:39:03 <jlinton> The flock presentation had a significant number of armv7 installs.
15:39:07 <pbrobinson> it is being discussed but there's still some interest in it
15:39:13 <nirik> I think we would want to give it a long notice... ie, not just decide and next release... but think a few ahead...
15:39:31 <pbrobinson> right and I've been thinking ahead already about some of this
15:40:03 <pbrobinson> in the general community there's still a lot of arm32 devices around that get used
15:40:07 <nirik> ok, fair enough. Just wanted to get a sense of usage
15:40:14 <pbrobinson> and they're *cheap*
15:40:34 <jlinton> I personally think the lift on armv7 is going to get greater since !debian has basically dropped all 32-bit targets (eg x86).
15:40:41 <pbrobinson> I've been trying to work out why the memory issue is a problem more of late
15:41:25 <pbrobinson> and I wonder if it's some of the compiler/toolchain features like LTO which put more pressure on memory when building, which is esp a problem on the single process limit
15:43:27 <nirik> could be
15:44:50 <pwhalen> Ok, lets hope the downgrade goes well for now and continue to discuss the plans long term.
15:44:59 <pwhalen> Anything else for today?
15:47:16 <pwhalen> #endmeeting