15:00:29 <pwhalen> #startmeeting Fedora ARM and AArch64 Status Meeting 15:00:29 <zodbot> Meeting started Tue May 24 15:00:29 2016 UTC. The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:29 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:29 <zodbot> The meeting name has been set to 'fedora_arm_and_aarch64_status_meeting' 15:00:29 <pwhalen> #chair pwhalen pbrobinson dgilmore hrw dmarlin yselkowitz jonmasters ahs3 masta 15:00:29 <zodbot> Current chairs: ahs3 dgilmore dmarlin hrw jonmasters masta pbrobinson pwhalen yselkowitz 15:00:39 * pbrobinson o/ 15:00:53 <pwhalen> morning folks, whos here today? 15:01:02 <yselkowitz> .hello yselkowitz 15:01:03 <zodbot> yselkowitz: yselkowitz 'Yaakov Selkowitz' <yselkowi@redhat.com> 15:01:07 * coremodule is here. 15:03:01 <pwhalen> #topic 1) ==== Package Status & Issues ==== 15:03:49 <pbrobinson> perl 5.24 has landed in rawhide 15:04:08 <pwhalen> #info Perl 5.24 has landed in rawhide. 15:04:12 <pbrobinson> so there's a bit of a mess there as it sorts itself out, should be done in the next day or two 15:04:34 <pbrobinson> the grub2/shim issue is closed 15:04:39 <pwhalen> \o/ 15:04:47 <pwhalen> anything new popping up? 15:05:06 <pbrobinson> we're freezing in a week, so I'll have a look through f24 to see if there's anything new but it all looks reasonable 15:05:47 <pbrobinson> yselkowitz: are you aware of anything that needs attention for f24? 15:05:53 <pwhalen> #info Final Freeze - 2016-05-31. 15:06:19 <yselkowitz> pbrobinson, not particularly 15:06:57 <pbrobinson> hrw: you about? I believe the libvirt stuff for the tianocore module has been fixed, did you test it? 15:07:22 <pbrobinson> other than that I have nothing to note 15:07:24 * masta looks in 15:07:27 <masta> hey gang 15:07:43 <pwhalen> yea, same, looks pretty good. 15:07:49 * pwhalen knocks on wood 15:07:55 <pwhalen> howdy masta 15:08:25 <pwhalen> ok, moving on.. 15:08:28 <pwhalen> #topic 2) ==== Kernel Status ==== 15:09:07 <pbrobinson> there's a few minor bits here, a couple of minor usb issues on imx6/tegra 15:09:23 <pwhalen> we've got a couple issues for armhfp, wandboard panic on shutdown and usb on tegra.. but overall looks good 15:09:26 <pbrobinson> I'm hoping to get a little polish on some of the aarch64 devices this week in 4.5 too 15:09:45 <masta> is there a bz for the tegra usb thing? 15:10:05 <pbrobinson> masta: not currently, it's fixed in 4.6, just need to work out a commit to pull back 15:10:14 <masta> ah! cool 15:11:11 <pbrobinson> 4.6 is currently mostly static, I'm sort of tracking anything we'll need there for when it goes to f24 but overall it looks reasonably good 15:11:22 <pbrobinson> and of course 4.7 is in full swing in rawhide at the moment 15:11:42 <pbrobinson> I've not looked closely at that yet 15:11:54 <pwhalen> ive not done too much with 4.7, few basic boot tests but thats ity 15:12:30 <pbrobinson> I'm mostly waiting there until rc1 in a week to assess where we are 15:13:21 <pwhalen> ok, other than polish/enablement for aarch64, has anyone seen issues we should be concerned about? 15:13:33 <pbrobinson> but over all I think for f24 on both armv7/aarch64 we're looking in areasonable state 15:13:46 <pbrobinson> anyone tested anything 15:14:01 <pbrobinson> any issues to report, any devices look good? 15:14:42 <pwhalen> ive been trying to get the firefly (rockchip) working with our uboot, but so far no success 15:15:14 <coremodule> I tried a kernel update on a RasPi 3 and, while the kernel *seemed* to correctly install, the output of "uname -a" still showed the old kernel. Did I do something wrong or has anyone else seen this? 15:15:40 <pbrobinson> coremodule: you do need to reboot, I'm not sure what image you're using, 15:16:14 <pbrobinson> coremodule: if it's a hacked fedora it might not update the boot process correctly for a new kernel so sorry your mostly on your own there atm 15:16:34 <masta> probably the boot scripts didn't update properly, or whatever 15:16:34 <coremodule> I did reboot, and it was a hacked Fedora, so alright, I gotcha. 15:16:45 <coremodule> Someone did mention something about changing the name of the kernel in /boot...? 15:17:23 <pbrobinson> coremodule: you might be able to update the config.txt with a new kernel name but I'm not exactly sure how the image you've created works in that regard 15:17:39 <coremodule> Gotcha, alright, just throwing it out there in case it was a common issue. 15:17:41 <masta> can be worked out in #fedora-arm 15:17:49 <coremodule> masta, Sure thing. 15:18:08 * jonmasters_ is in but on a call as well 15:18:29 <pbrobinson> coremodule: no common issue here 15:18:43 <pwhalen> ok, so kernel looks pretty good.. if nothing else lets move on .. 15:18:51 <masta> what needs testing these days? 15:18:58 <masta> (what boards) 15:19:01 <pbrobinson> masta: what have you got to test 15:19:36 <masta> pbrobinson, I got a box of boards here, packed when I moved a year ago to new address. Guess I'll just start loading rawhide on them and see how it goes 15:19:43 <pwhalen> masta, i think you have a bunch of things i dont, so whatever youve got.. i update the support hw list with what ive tested 15:19:48 <pbrobinson> masta: start with f24 please 15:19:50 <masta> otay 15:20:00 <pwhalen> well, f24 for now.. 15:20:01 <pbrobinson> masta: I think you had a omap5 device that would be good to test 15:20:21 <masta> pbrobinson, ah! f24 is fine by me, but I'd like to get the latest kernel tested too ;-) 15:20:34 <pbrobinson> masta: but when you bother to unpack the box sync on #fedora-arm and we can assist in where the help would be useful 15:20:47 <pwhalen> masta, http://fedoraproject.org/wiki/Architectures/ARM/F24/Installation/Hardware_Status 15:20:54 <pbrobinson> masta: start with f24 then you can add new kernel/u-boot etc 15:21:33 <pwhalen> so anything not on there, or missing results would be great 15:22:00 <masta> awesome thanks 15:22:33 <pwhalen> #topic 3) ==== Bootloader Status ==== 15:23:09 <pbrobinson> so overall I think grub2/shim issue is resolved 15:23:42 <pbrobinson> there's a BZ with a gigabyte board that is not verified but it's hard to tell from the reports 15:23:46 <pwhalen> i tested last week on seattle, qemu and mustang. looked good 15:24:07 <pbrobinson> excellent 15:24:11 * pwhalen will update bz 15:24:46 <pbrobinson> on the u-boot front I think 2016.05 is looking good now on all the platforms I've tested on 15:24:59 <pwhalen> #info Latest - uboot-tools-2016.05-3.fc25 15:25:04 <pbrobinson> the question is whether we want to look at landing it in f-24 or not 15:25:05 <pwhalen> #info Added Rockchip SPL. Fix distro boot on clearfog. arm64 EFI boot fixes 15:25:15 <pwhalen> #link http://koji.fedoraproject.org/koji/buildinfo?buildID=767838 15:25:21 <masta> it would have to be soon 15:25:32 <pwhalen> so, we need to decide that this week 15:25:42 <pbrobinson> masta: that's the statement of the bleeding obvious 15:26:06 <masta> pbrobinson, that is my specialty ;-) 15:26:21 <pwhalen> it did look pretty good, we need to check everything this week and discuss next week? 15:26:48 <pbrobinson> no we need to discuss early this week and land friday at the latest to ensure karma to land in before freeze 15:27:50 <pbrobinson> so who has tested 2016.05 other than pwhalen and I 15:28:34 <masta> hehe... guess it's just you two 15:29:02 <pwhalen> masta et al, could you try out what you got this week? let us know 15:29:03 <coremodule> Yeah, not me. 15:29:34 <pwhalen> be awesome, else we'll make a decision based on what we have, which might not work on your board(s) 15:29:55 <pwhalen> thanks guys, appresh 15:30:18 <masta> sure, I'll probably try a few boards starting Wednesday. 15:30:40 <pwhalen> #topic 4) ==== F24 Final ==== 15:30:57 <pbrobinson> I think we should have a sync around this time tomorrow on #fedora-arm and see how we're looking there 15:31:04 <pwhalen> yesterday was a holiday here, so i missed the blocker meeting 15:31:13 <pbrobinson> I think overall we're looking OK 15:31:42 <pwhalen> #link https://qa.fedoraproject.org/blockerbugs/milestone/24/final/buglist 15:31:55 <pbrobinson> the last thing I have on my todo list is aarch64 disk images for SBCs, I can now build them, I'm just trying to verify them booting on a couple of devices 15:32:22 <pbrobinson> we have docker and cloud images (and firmware/libvirt etc support for cloud images) in place 15:32:34 <pbrobinson> so I think over all we're looking OK 15:33:11 <masta> pbrobinson, livemedia-creator? 15:33:18 <pbrobinson> masta: nope 15:33:32 <pbrobinson> anyone else aware of any issues? 15:33:46 <coremodule> masta, livemedia-creator is looking good... 15:33:58 <pbrobinson> coremodule: in what context? 15:34:29 <coremodule> It correctly takes an .iso and loads it to disk/USB bootably. 15:34:37 <coremodule> And reproducibly. 15:34:54 <pbrobinson> coremodule: wrong context, it's in koji builds for aarch64 that we need and it's not yet supported 15:35:04 <coremodule> pbrobinson, Ah, sorry. 15:35:08 <masta> yep 15:35:12 <pwhalen> #topic 5) == Open Floor == 15:35:27 <pwhalen> we're in the final stretch, anything else for today folks? 15:35:57 <yselkowitz> at some point we may need some help with https://bugzilla.redhat.com/show_bug.cgi?id=1289389 15:36:11 <masta> had a surprising amount of aarch64 questions at the OSCON Fedora booth 15:36:20 <yselkowitz> I think that's a bug in python itself 15:36:33 <pbrobinson> masta: how did you answer them given you've basically not been around? 15:37:10 <masta> pbrobinson, hehe... 15:37:14 <pwhalen> awsome, what sort of questions? 15:37:40 <pbrobinson> yselkowitz: when you file aarch64 bugs can you _PLEASE_ link them against ARM64 blocker bug so someone knows they actually exist 15:39:14 <yselkowitz> it is blocking, did you just add that? 15:39:44 <pbrobinson> yselkowitz: no, it's in some random epel7aarch64 bug, yet this is reported against f24 15:40:09 <pbrobinson> yselkowitz: yet the bug it depends on is epel 15:40:20 <pbrobinson> or RHELSA 15:40:20 <yselkowitz> there are two bugs, that's the other one 15:40:33 <yselkowitz> the fedora bug is https://bugzilla.redhat.com/show_bug.cgi?id=1289389 15:41:14 <pbrobinson> yselkowitz: yet they are cross linked in the blocks 15:41:41 <yselkowitz> the fedora bug blocks ARM64 15:41:51 <yselkowitz> the epel bug blocks epel7aarch64 15:42:24 <pbrobinson> yselkowitz: yes but the fedora bug also depends on the bug reported against rhelsa, so they look mixed, I'll look at it later 15:42:47 <pwhalen> anything else for open floor today? 15:43:47 <pwhalen> #endmeeting