15:00:37 <pwhalen> #startmeeting Fedora ARM and AArch64 Status Meeting 15:00:37 <pwhalen> #chair pwhalen pbrobinson coremodule jlinton 15:00:37 <zodbot> Meeting started Tue Apr 12 15:00:37 2022 UTC. 15:00:37 <zodbot> This meeting is logged and archived in a public location. 15:00:37 <zodbot> The chair is pwhalen. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 15:00:37 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:37 <zodbot> The meeting name has been set to 'fedora_arm_and_aarch64_status_meeting' 15:00:37 <zodbot> Current chairs: coremodule jlinton pbrobinson pwhalen 15:00:37 <pwhalen> #topic roll call 15:01:01 <pwhalen> Good morning folks, who's here today? 15:01:25 <coremodule> Hi pwhalen, I'm here 15:01:28 * jlinton waves 15:03:32 <pwhalen> Morning gents 15:03:39 <pwhalen> #topic 1) ==== Userspace Status ==== 15:03:54 <pwhalen> Any new issues to discuss? 15:04:17 <pwhalen> I'm not aware of anything myself 15:04:36 <pwhalen> other than F36 related, of course 15:04:49 <pwhalen> #info No new issues reported. 15:05:15 <pwhalen> #topic 2) ==== Kernel Status ==== 15:05:27 <pwhalen> #info kernel-5.17.2-300.fc36 15:05:27 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1944278 15:06:04 <pwhalen> thankfully, this 'fixes' two issues for us, but it sounds like we have gotten lucky on one of them 15:06:18 <pwhalen> #info RPi4 wired network fails - eth0 (bcmgenet): transmit queue 1 timed out 15:06:18 <pwhalen> #link https://bugzilla.redhat.com/show_bug.cgi?id=2053729 15:06:45 <jlinton> Yah, I need to post a revert, so unless the compiler lands first, we will need to carry an out of tree patch to keep it working. 15:06:57 <pwhalen> This was because of jlintons patch, right.. 15:07:23 <pwhalen> #info early kernel panic on aarch64 with kernel-5.17.0-0.rc2.20220202+ 15:07:23 <pwhalen> #link https://bugzilla.redhat.com/show_bug.cgi?id=2053214 15:08:12 <jforbes_eh> Is a new version of gcc even being built with that patch soonish? 15:08:21 <pwhalen> The seattle and mustang both boot with 5.17.2, but it sounds like we got lucky 15:08:46 <pwhalen> jforbes_eh: no idea myself, I hope so 15:09:03 <pwhalen> jforbes_eh: Are you Canadian? :) 15:09:29 <jlinton> Yah, i'm upgrading my seattle as we speak 15:09:36 <jforbes_eh> Sorry for the nick, the ZNC server I typically use is down and I didn't have a non ZNC setup for libera yet 15:10:08 <pwhalen> heh, :) 15:10:34 <pwhalen> jlinton: I tested device tree and meant to do acpi but got distracted, will do after the meeting 15:10:57 <jforbes_eh> Reason I was asking, is 5.17.3 is out for review now, so I will be building it in a couple of days, could wait if it was going to be a day or so, but not too much longer 15:11:04 <jlinton> Well, thats good because I just about never test DT 15:11:11 <pwhalen> After booting successfully, I did fresh f36 installations on both 15:11:26 <pwhalen> its our default, so I tend to check it first 15:12:45 <pwhalen> I'm happy 5.17.2 works for the most part, so at least my local legacy hardware will still work with the release 15:12:48 <jforbes_eh> Speaking of, my NX acpi data might be incorrect, but I can't verify until later today since my USB serial cable disappeared. New one arriving in a few hours 15:13:29 <pwhalen> Pretty sure its still broken for acpi 15:13:59 <pwhalen> Any other kernel issues to discuss? 15:14:25 <jforbes_eh> None that I have seen updates on recently 15:14:29 <jlinton> so 5.17.2 just booted on the seattle too, so we either got lucky, or someone fixed something that helps it out. 15:14:48 <pwhalen> nice, thanks for confirming. 15:15:03 <pwhalen> #info Please test and report any issues to the list or #fedora-arm. 15:15:26 <pwhalen> #topic 3) ==== Bootloader Status ==== 15:16:11 <pwhalen> No change here, nor did I see any problems reported. 15:16:17 <pwhalen> #info uboot-tools-2022.04-1.fc36 15:16:17 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1942618 15:17:11 <pwhalen> I tested a bunch of armhfp devices, couple aarch64 15:17:22 <pwhalen> coremodule: did you test the nano by chance? 15:17:33 <coremodule> Yeah, but no joy here... 15:18:01 <coremodule> I haven't had time to look into it further, it's been working for other people, so once the F36 cycle is complete, I'll figure out where it went wrong. 15:18:08 <pwhalen> no joy meaning it didnt boot? 15:18:22 <coremodule> it boots, just no video out for me. I can use the serial console to do everything else. 15:18:47 <coremodule> If I run F34, everything is fine, so there's some regression between then and now. 15:19:16 <coremodule> If I run F34, video out works I should say. 15:19:55 <pwhalen> Right, and I've seen others report the same in f36.. but it works for me most of the time 15:20:30 <coremodule> Oh! Really? Others are having this issue? I can't *ever* get video out... And I've tried probably around 50 times total. 15:21:29 <pwhalen> Yes, can't recall who though.. maybe in channel 15:22:01 <jlinton> On the efi framebuffer? Cause something looks to have changed in libvirt/etc as well, I wasn't seeing the tianocore graphics+grub on the framebuffer, only on the serial 15:22:36 <jlinton> (although in that case its fine once its booted) 15:23:18 <pwhalen> I think that was reported already, not following it too closely though 15:23:24 <jforbes_eh> Wonder if the simpledrm stuff is causing a problem for f36? 15:24:15 <jforbes_eh> If so, we can test soon. I am doing 5.17 builds for F35 and F34 starting with 5.17.3 which have the simpledrm reverted. I can do a one off F36 build to test. 15:24:37 <jforbes_eh> I haven't tried f36 on the nano yet though 15:25:11 <coremodule> jforbes_eh, that would be good. I can give it a test on my nano here and see what happens... 15:26:18 <pwhalen> thanks both 15:26:24 <pwhalen> #topic 4) ==== Fedora 36 ==== 15:26:53 <pwhalen> #info Current testing results: 15:26:53 <pwhalen> #link https://fedoraproject.org/wiki/Test_Results:Fedora_36_Branched_20220405.n.0_Summary 15:27:47 <jlinton> Yah, I've tested it on rpi4/acpi, ampere altra, seattle (now boots), thunderX2, mcbin, and honeycomb and they boot, networking works, etc 15:27:57 <pwhalen> #info List of Blockers for Fedora Linux 36 (Final) 15:27:57 <pwhalen> #link https://qa.fedoraproject.org/blockerbugs/milestone/36/final/buglist 15:28:17 <pwhalen> thanks jlinton, I haven't tested enterprise in a while 15:28:31 <jlinton> Yah, those are all in ACPI mode. 15:29:19 <pwhalen> The list of blockers looks long. coremodule any idea where we are for an RC? Do you think we're slippin'? 15:29:27 <jforbes_eh> I have tested on the NX in DT mode 15:29:55 <pwhalen> thanks jforbes_eh 15:30:11 <pwhalen> do you install to nvme or? 15:30:26 <jforbes_eh> I did a card run and an NVME 15:30:46 <pwhalen> cool, thanks 15:30:47 <jforbes_eh> Testing image to card seemed prudent before blowing away a working nvme :) 15:30:54 <pwhalen> heh 15:30:55 <coremodule> pwhalen, yeah, we're slippin'... Like you said, no RC... 15:31:37 <pwhalen> Right, but its only tuesday.. is Adam out this week? 15:31:53 <coremodule> Yeah, he's out until next Monday. 15:32:34 <coremodule> Well, he's traveling until next Monday, he might be in and out, but not full time. 15:32:45 <pwhalen> Ok. 15:32:56 <pwhalen> Anything else for F36? 15:34:07 <pwhalen> #topic 5) == Open Floor == 15:34:23 <pwhalen> Anything for open floor today? 15:36:34 <pwhalen> #endmeeting