15:00:45 <coremodule> #startmeeting Fedora ARM Working Group Meeting
15:00:45 <zodbot> Meeting started Tue Jun 13 15:00:45 2023 UTC.
15:00:45 <zodbot> This meeting is logged and archived in a public location.
15:00:45 <zodbot> The chair is coremodule. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
15:00:45 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:45 <zodbot> The meeting name has been set to 'fedora_arm_working_group_meeting'
15:00:45 <coremodule> #chair pwhalen pbrobinson jlinton coremodule
15:00:45 <zodbot> Current chairs: coremodule jlinton pbrobinson pwhalen
15:00:45 <coremodule> #topic roll call
15:00:49 <coremodule> Hi all, who is around today?
15:01:31 <jforbes> morning
15:01:45 <coremodule> #chair jforbes
15:01:45 <zodbot> Current chairs: coremodule jforbes jlinton pbrobinson pwhalen
15:01:55 <jlinton> morning
15:04:04 <coremodule> lets get started!
15:04:05 <coremodule> #info Fedora-Cloud-38-20230523.0
15:04:05 <coremodule> #link https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=37&build=Fedora-Cloud-37-20230613.0&groupid=1
15:04:21 * pwhalen is here
15:04:37 <coremodule> anything for f38 today? I don't have anything
15:05:34 <coremodule> #undo
15:05:34 <zodbot> Removing item from minutes: <MeetBot.items.Link object at 0x7f1e01880f28>
15:05:35 <jlinton> Not me
15:05:37 <coremodule> #undo
15:05:37 <zodbot> Removing item from minutes: INFO by coremodule at 15:04:05 : Fedora-Cloud-38-20230523.0
15:05:46 <coremodule> #topic 1) ==== Stable Releases ====
15:05:46 <coremodule> #info Fedora-Cloud-38-20230523.0
15:05:46 <coremodule> #link https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=37&build=Fedora-Cloud-37-20230613.0&groupid=1
15:05:51 <coremodule> there we go, that's better
15:06:03 <coremodule> alright, we can jump to the kernel then
15:06:04 <coremodule> #topic 2) ==== Kernel Status ====
15:06:04 <coremodule> #info F38 kernel-6.3.7-200.fc38
15:06:04 <coremodule> #info https://koji.fedoraproject.org/koji/buildinfo?buildID=2211324
15:06:04 <coremodule> #info Rawhide kernel-6.4.0-0.rc6.48.fc39
15:06:05 <coremodule> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=2212294
15:06:27 <coremodule> I think I'm going to try this format, that way we get the latest stable kernel and the latest rawhide kernel shown
15:06:38 <jforbes> The 6.3 kernels should be stabilized at this point, I don't see any major arm specific issues
15:07:24 <jforbes> 6.4 is also nearing rc6, so it is a good time to start thinking about what new drivers shoudl be flipped on based on what hardware is actually out there and potentially useful for Fedora
15:08:51 <jforbes> nearing, we are at rc6, so expecting 2 weeks before .0
15:09:38 <jforbes> nothing else kernel specific at this time
15:09:53 <coremodule> I think pbrobinson would know best which hardware to keep/use
15:10:07 <coremodule> jforbes, is there a kernel-specific meeting each week?
15:11:08 <jforbes> coremodule: there is not. We used to run them, many years ago, but they were found to be of little use
15:11:45 <jforbes> And particularly as the kernel team isn't so much anymore, just ping me in IRC any time :)
15:12:00 <coremodule> okay, gotcha. thanks for the info!
15:12:07 <SumantroMukherje> Jforbes, do you think it will be a good thing to do?
15:12:28 <jforbes> I do not think there would be much value to it
15:12:49 * SumantroMukherje nods
15:13:10 <jforbes> The biggest area of churn which requires some discussion is really arm hardware enablement. pbrobinson is usually on top of that
15:13:17 <jlinton> A better discussion might be what to turn of
15:13:19 <jlinton> off
15:13:39 <jlinton> since there are likely a lot of 32-bit boards that are no longer supported that still have clk/etc drivers statically bound.
15:14:01 <jlinton> I was going to go through it a few months ago and see what I could kill off, but pbrobinson mentioned something similar, but I don't think its been done.
15:14:06 <jforbes> Well, we did turn off armv7, and we should probably turn off anything specific to ThunderX first gen stiff
15:14:16 <jforbes> err stuff
15:14:35 <jforbes> As that original ThunderX stuff is expected to be broken with 6.4+
15:15:26 <jforbes> But yes, we are much better at turning on things than we are at turning off.
15:15:32 <jlinton> except i'm not sure how many things specific to tx1 are still being used by newer platforms.
15:16:05 <jlinton> well because its obvious when something isn't working, less obvious when something is on that is no longer in use.
15:16:48 <jforbes> That, and even things that I assume no one could possibly still be using, I turn off and find someone who complains
15:16:53 <jlinton> But there were things I saw during bootup with some additional debug that should be off.
15:17:52 <jlinton> I added a print to the static module init callback, and I was guessing about 1/2 of the items weren't needed anymore.
15:18:19 <jforbes> MRs for config changes are gladly accepted
15:18:47 <jlinton> Yah, i've got a bit of breathing space unless I go after some new bugs, so let me repeat that process and see what I can disable.
15:19:01 <jlinton> the idea originally was to slim down the kernel image itself.
15:19:19 <coremodule> jforbes, would it be helpful to schedule something in this meeting specifically for enablement for next week? like send an email? or do you usually contact those who would know best one-on-one?
15:19:33 <jforbes> It also shrinks the attack surface in some cases
15:19:45 <jforbes> coremodule: pbrobinson usually just sends an MR
15:20:05 <coremodule> gotcha, okay. I'm just trying to make sure you get what you need, at least from the arm team
15:20:16 <jforbes> I only brought it up here, because if there are things that people in the community need, it is a good time to ask
15:21:16 <jforbes> there are platforms that we may not "support" but that people are working with, trying to get working. I don't mind turning on drivers for that purpose
15:22:07 <jlinton> Some of these DPU's might be a good sanity check, although I don't personally have access to any.
15:22:10 <coremodule> that makes sense
15:22:54 <jlinton> maybe BF2, but that should already work.
15:23:21 <jlinton> BF3/etc probably need attention.
15:24:54 <coremodule> alright, thanks jforbes
15:25:02 <coremodule> lets move to the bootloader
15:25:03 <coremodule> #topic 3) ==== Bootloader Status ====
15:25:03 <coremodule> #info uboot-tools-2023.07-0.3.rc4.fc39
15:25:04 <coremodule> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=2212359
15:26:05 <coremodule> #info New uboot-tools released yesterday, with fixes for the Pinephone Pro, RockPro64.
15:26:25 <coremodule> that's all I have here, testing would be appreciated
15:27:24 <coremodule> #topic 4) ==== Fedora 39 status ===
15:27:24 <coremodule> #info OpenQA Testing Fedora-Rawhide-20230606.n.0
15:27:24 <coremodule> #info https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=Rawhide&build=Fedora-Rawhide-20230613.n.0&groupid=5
15:28:17 <coremodule> We're just under two months from branch point. So far, rawhide has looked good. I don't have anything specific here either.
15:31:49 <coremodule> #topic 5) ==== Open Floor ====
15:33:06 <jforbes> Just a small update on the X13s.
15:33:52 <jforbes> There is an MR against dracut to make it bring the ICC drivers into the initramfs, should be a big help.  Once that is done, we should be able to get a stock rawhide image to boot
15:34:10 <jforbes> Some of the graphics work isn't expected to land until 6.5 though
15:34:34 <jforbes> That is all of interest
15:37:52 <coremodule> thanks jforbes. Is 6.5 supposed to ship with f39 or are we that far along to know yet?
15:38:52 <jforbes> I expect it will
15:39:38 <jforbes> Or at least that is the current plan
15:40:34 <coremodule> cool, good to know
15:40:46 <coremodule> alright, anything else for open floor or should we end this thing?
15:42:19 <coremodule> thanks everyone
15:42:22 <coremodule> #endmeeting