fedora_arm_and_aarch64_status_meeting
LOGS
15:00:18 <pwhalen> #startmeeting Fedora ARM and AArch64 Status Meeting
15:00:18 <zodbot> Meeting started Tue Aug 18 15:00:18 2020 UTC.
15:00:18 <zodbot> This meeting is logged and archived in a public location.
15:00:18 <zodbot> The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:18 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:18 <zodbot> The meeting name has been set to 'fedora_arm_and_aarch64_status_meeting'
15:00:18 <pwhalen> #chair pwhalen pbrobinson jlinton
15:00:18 <pwhalen> #topic roll call
15:00:18 <zodbot> Current chairs: jlinton pbrobinson pwhalen
15:00:35 <pwhalen> Good morning folks, who's here today?
15:01:43 * pbrobinson o/
15:02:24 * jlinton waves
15:02:53 <pwhalen> Good morning gents, lets get started
15:03:00 <pwhalen> #topic 1) ==== Userspace Status  ====
15:03:28 <pwhalen> Any new user space issues in the stable releases?
15:04:13 <pwhalen> none come to mind
15:04:30 <pwhalen> #info No new issues reported.
15:04:40 <pwhalen> #topic 2) ==== Kernel Status ====
15:04:55 <pwhalen> #info kernel-5.8.1-300.fc32 coming to F32..
15:04:55 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1592522
15:05:32 <pwhalen> Not sure the time line, but 5.8 will hit F32 soon.
15:05:49 <jlinton> I believe that the DMA stuff in 5.8 should be ok on rpi4, but I will need to check the fedora build
15:06:52 <pwhalen> #info Kernel 5.8 will be the GA kernel for F33 final, 5.9 to be an update to be available shortly after F33 is released.
15:07:14 <pwhalen> thanks jlinton
15:07:41 <pwhalen> #info kernel-5.9.0-0.rc1.1.fc34
15:07:41 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1594084
15:08:04 <pwhalen> I just noticed 5.9 rc1 before the meeting
15:08:13 <pwhalen> #info Please test and report any issues to the list or #fedora-arm.
15:09:58 <pwhalen> Any other kernel news, or issues?
15:10:44 <pwhalen> #topic 3) ==== Bootloader Status ====
15:11:05 <pwhalen> #info Latest U-Boot available for testing -  uboot-tools-2020.10-0.2.rc2.fc33
15:11:05 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1588325
15:12:47 <pwhalen> There is also a new grub2 build which breaks pxe boot, I'll cover that in F33 next.
15:13:33 <pwhalen> Anything else to mention for bootloaders?
15:15:46 <pbrobinson> jlinton: it boots but it doesn't fix usb
15:16:39 <pbrobinson> there also looks to be issues, for me at least, with the u-boot 2020.10 builds on some aarch64 devices but I need to dig further
15:17:04 <pwhalen> pbrobinson: what's happening?
15:17:15 <pbrobinson> it's crashing and reseting
15:17:20 <pbrobinson> not sure if it's just me though
15:17:35 <pbrobinson> I did see 1 report on the 8gb rpi4 on iot
15:17:55 <pwhalen> I need to test the latest rc, ok. Thanks.
15:18:09 <pwhalen> #topic 4) ==== Fedora 33 ====
15:18:28 <pwhalen> #info Fedora 33 has now branched from Rawhide.
15:18:45 <pwhalen> #link https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
15:19:25 <pwhalen> #info List of Blockers - https://qa.fedoraproject.org/blockerbugs/milestone/33/beta/buglist
15:19:49 <pwhalen> We got a new version of grub2 over the weekend, causes issues with pxe boots
15:20:01 <pwhalen> #info error: ../../grub-core/net/net.c:1795:timeout reading initrd.img
15:20:01 <pwhalen> #link https://bugzilla.redhat.com/show_bug.cgi?id=1869335
15:20:18 <pwhalen> #info If affected use grub2-2.04-25.fc33 until a new build is available.
15:20:29 <pwhalen> that is also an accepted blocker.
15:20:33 <jlinton> Everywhere I look I see pac/bti sequences, so that good, and I've running a recent compose on a 8.0 machine without any serious issues yet
15:20:49 <jlinton> OTOH, elfutils seems to have grown a few more problems
15:20:55 <jlinton> in its unit tests
15:21:12 <jlinton> so i'm looking at those (along with the idea that I want the pac/bti progbits to be more visible)
15:21:31 <pwhalen> excellent, thanks jlinton
15:21:35 <pwhalen> jlinton++
15:21:36 <jlinton> so we can run readelf and see right away if the pac/bti flags are set
15:23:00 <jlinton> Steve C, has a global disable patch for the kernel command line, that we should still be looking at. I expect to be able to spin it up on a ARM v8.4 machine RSN (not a model).
15:24:30 <pbrobinson> cool
15:24:33 <pwhalen> sounds good
15:24:57 <pwhalen> we've has an issue with armhfp installs again..
15:25:08 <pwhalen> #info zram-setup@zram0.service: Failed to load configuration: No such file or directory
15:25:08 <pwhalen> #link https://bugzilla.redhat.com/show_bug.cgi?id=1861463
15:25:17 <pwhalen> #info This is now fixed with changes made to the udisks2-zram package.
15:25:20 <pbrobinson> jlinton: were you also going to look at the grub2 on armv7 issue?
15:25:36 <jlinton> Yah, I have that on my list... sadly no progress.
15:26:37 <jlinton> OTOH, I was under the impression it was less a grub problem, and more an issue with how the install/compose is run.
15:27:58 <pbrobinson> jlinton: want to extrapolate there?
15:28:01 <jlinton> (or put another way, how the efi system partition is being created/mounted during the compose, although some of that might be how grub-install is run)
15:28:09 <pbrobinson> it should be working in the same way as aarch64
15:28:41 <jlinton> yah, that might be part of my confusion, I went looking for special cases (well a few weeks ago) and didn't see anything, so I wasn't sure if I was actually looking at the right scripts
15:31:06 <jlinton> I guess it could be something wrong with the way grub determines if its on a BIOS vs EFI machine, maybe something is failing to detect EFI.
15:31:13 <jlinton> This I guess is an offline conversation..
15:32:01 <pwhalen> we can continue that in channel after the meeting. Would be great to have working in F33.
15:32:06 <pwhalen> #info Creating an armhfp disk image with btrfs results in an empty filesystem
15:32:06 <pwhalen> #link https://bugzilla.redhat.com/show_bug.cgi?id=1855034
15:33:17 <pwhalen> This one is still an issue, there were some changes made but its still not sufficient and we're running out of time. I'll propose this as a blocker.
15:33:38 <pbrobinson> jlinton: that may be imagefactory
15:34:23 <pwhalen> #info Fedora 33 Workstation Live ISO for AArch64 now available for testing
15:34:23 <pwhalen> #link https://dl.fedoraproject.org/pub/fedora-secondary/development/33/Workstation/aarch64/iso/
15:35:20 <pwhalen> #info F33 Beta Freeze - starts at 1400 UTC on 2020-08-25
15:35:37 <pwhalen> #info F33 Beta Target Release - 2020-09-22
15:36:30 <pwhalen> Anything else for F33?
15:38:12 <pwhalen> #topic 5)  == Open Floor ==
15:39:05 <pwhalen> Anything else for today?
15:39:27 <pbrobinson> nope
15:40:18 <pwhalen> #endmeeting