16:01:42 <pwhalen> #startmeeting Fedora ARM and AArch64 Status Meeting 16:01:42 <zodbot> Meeting started Tue Mar 8 16:01:42 2016 UTC. The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:42 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:01:42 <zodbot> The meeting name has been set to 'fedora_arm_and_aarch64_status_meeting' 16:01:51 <pwhalen> hrw, thanks, was distracted :) 16:01:55 <pwhalen> whos here today? 16:02:00 <hrw> o/ 16:02:07 <hrw> #chair them 16:02:49 <pwhalen> #chair pwhalen bconoboy pbrobinson dgilmore hrw jsmith dmarlin 16:02:49 <zodbot> Current chairs: bconoboy dgilmore dmarlin hrw jsmith pbrobinson pwhalen 16:03:29 <dgilmore> hi 16:03:31 <pwhalen> we'll give it a few minutes, see if others join us 16:03:34 <pwhalen> morning dgilmore 16:04:25 <masta> hey guys 16:04:41 <pwhalen> morning masta 16:04:55 <masta> morning pwhalen :-) 16:08:38 <pwhalen> hrm, we have a number of things to discuss, but doesnt look like we have the a quorum here today, thoguhts? 16:09:19 <hrw> I would go for it anyway 16:09:45 <hrw> it will be short 16:09:58 <pwhalen> ok 16:10:01 <pwhalen> #topic 1) ==== Package Status & Issues ==== 16:10:36 <pwhalen> pbrobinson, shared some stats this morning for f24 aarch64: 16:10:36 <pwhalen> statistics: {'older': 156, 'local_only': 1, 'remote_only': 322, 'same': 17535, 'newer': 0, 'total_missing_builds': 250} 16:10:46 * pbrobinson waves, sorry 16:10:58 <pbrobinson> so no major issues at the moment 16:11:00 <pwhalen> my understanding is we're done the mass rebuild... 16:11:01 <pbrobinson> mass rebuild is done 16:11:04 <pwhalen> nice! 16:11:08 <pbrobinson> now just cleanups all around 16:11:27 <hrw> I went with qt5-qtwebengine again. altered libvpx spec and looks like it (qt-qtw) builds now. 16:11:29 <pwhalen> #info Mass rebuild for F24 AArch64 complete, now time for cleanups. 16:11:33 <pbrobinson> issue with new grub2 build, working with the maintainer to fix it now (that's why I was late here) 16:11:37 <pwhalen> #info statistics: {'older': 156, 'local_only': 1, 'remote_only': 322, 'same': 17535, 'newer': 0, 'total_missing_builds': 250} 16:12:09 <pwhalen> #info issue with new grub2 build, pbrobinson is working with the maintainer to fix it. 16:12:22 <pbrobinson> f25 is now building as well, about 1200 builds behind there, should be caught up in a couple of days 16:12:50 <pwhalen> #info f25 for AArch64 is now building as well, about 1200 builds behind there, should be caught up in a couple of days 16:13:20 <pbrobinson> there's a bunch of cleanups we'll need to do around individual package failure but there's nothing particularly that's hard blocking other builds (except maybe firefox/xulrunner) 16:13:35 <hrw> gecko ;( 16:13:38 <pbrobinson> but there's a new version incoming for that with 45 on it's way in 16:14:27 <pbrobinson> over all we're looking pretty good for pure builds as we run up to alpha 16:14:28 <hrw> pbrobinson: how we are with logs on arm.koji? still not available? 16:14:47 <pbrobinson> hrw: that's what I was looking at when I was late for the meeting 16:15:02 <hrw> pbrobinson: thanks a lot 16:15:09 <pbrobinson> it's in my very large list of things to fix 16:15:15 <pwhalen> numbers are pretty good, for comparison for f23 GA : {'older': 36, 'local_only': 0, 'remote_only': 292, 'same': 16752, 'newer': 0, 'total_missing_builds': 36} 16:15:16 <hrw> simplifies selecting which pacakge to look at 16:15:39 <hrw> pwhalen: nice 16:15:40 <pwhalen> as you can see, our 'same' numbers have increased quite a bit since f23 GA. 16:16:29 <pwhalen> anything else? 16:16:40 <pbrobinson> anyway, that's package failure... just a general overall for f24... 16:17:02 <pbrobinson> should be composing again for nightlies for f25 and f24 in the next day or so 16:17:14 <pbrobinson> just getting everything for the new compose process in place 16:17:20 <pbrobinson> most of the packages should be signed 16:17:39 <pwhalen> #info Nightlies for f25 and f24 AArch64 should start within the next couple of days 16:17:45 <pbrobinson> but also having some issues with signing at the moment, worked one day broke the next 16:18:10 <pwhalen> ive noticed signing issues all around, saw some discussion on it 16:18:28 <pbrobinson> plan is to initially get nightly Everything repo (the usual to what we've always had) plus nightly server and boot iso installers 16:19:02 <pbrobinson> then we'll add to that various other components like images for SBCs etc 16:19:04 <pwhalen> #info Packages may not be signed (all arches), currently being looked at and should be resolved shortly. 16:19:47 <pwhalen> sounds good. 16:20:16 <pwhalen> #topic 2) ==== Kernel Status ==== 16:20:23 <pwhalen> big win here. 16:20:47 <hrw> .bug 1303147 16:20:52 <zodbot> hrw: Bug 1303147 gcc 6 breaks booting on numerous ARM devices - https://bugzilla.redhat.com/1303147 16:21:01 <pwhalen> #info gcc 6 breaks booting on numerous ARM devices (bz#1303147) 16:21:07 <hrw> #link https://bugzilla.redhat.com/1303147 gcc 6 breaks booting on numerous ARM devices 16:21:09 <pbrobinson> so yes, hrw found us a work around for Cortex-A7/A15 devices not booting 16:21:17 <pwhalen> \o/ hrw 16:21:34 <pbrobinson> so that's enabled as of kernel-4.5.0-0.rc7.git0.2.fc24 16:21:38 <hrw> #link http://gcc.gnu.org/bugzilla/show_bug.cgi?id=70128 upstream gcc bug for it 16:22:00 <pwhalen> #info Kernels now booting on Cortex-A7/A15 devices 16:22:06 <hrw> looks like gcc guys started to wonder how it worked at all with gcc5 16:22:13 <pbrobinson> so all ARMv7 devices should now be in a generally working state again, just in time for freeze! 16:22:21 <pwhalen> tg 16:22:27 <hrw> debugging was fun 16:22:29 <pbrobinson> even earlier than that... 16:23:11 <pwhalen> hrw, didnt look that hard *after* you explained what you did, but tedious 16:23:32 <pwhalen> thanks again for that 16:23:32 <pbrobinson> so as far as I'm aware there's no general blocking issues now, so people need to test all devices as much as possible 16:23:37 <hrw> pwhalen: 32on64 helps 16:24:06 <hrw> https://bugzilla.redhat.com/show_bug.cgi?id=1309147 (initrd on aarch64) needs checking with newer kernel etc 16:24:24 <pwhalen> #info Please resume testing on all devices, report issues to the list/Bugzilla 16:25:05 <pbrobinson> hrw: interestingly the functionality in the kernel is supported on ARMv8 too yet they never enabled it for that class of devices 16:25:17 <hrw> ah, one more thing to add 16:25:25 <hrw> pbrobinson enabled new devices in kernel 16:25:31 <pwhalen> yea, thats the next big one to resolve, imo 16:25:43 <pwhalen> the unpack initrd 16:26:22 <pbrobinson> mostly aarch64 including PINE64 and Geekbox, they'll be easier to test soon once I get time to look at nightly images (as mentioned above) and other bits like bootloaders 16:26:48 <pwhalen> #info PINE64 and Geekbox now enabled as of 4.5.0-0.rc7.git0.1 16:26:52 <pbrobinson> and also Jetson TX1 for aarch64 16:27:08 <hrw> pbrobinson: r/pi[23] and tegra for arm32 iirc 16:27:17 <pbrobinson> hrw: nope 16:27:21 <pwhalen> #undo 16:27:21 <zodbot> Removing item from minutes: INFO by pwhalen at 16:26:48 : PINE64 and Geekbox now enabled as of 4.5.0-0.rc7.git0.1 16:27:26 * pbrobinson kicks hrw 16:27:42 <pwhalen> #info PINE64, Jetson TX1 and Geekbox now enabled as of 4.5.0-0.rc7.git0.1 16:28:31 <hrw> ah, right - tegra for jetson. 16:29:10 * hrw goes to git reset fedora-packager/kernel 16:29:25 <pwhalen> anything else to add to that long list? 16:29:27 <pwhalen> new kernel issues? 16:30:03 <hrw> nothing hit me during last week 16:30:03 <pbrobinson> none I'm aware of at the moment 16:30:08 <pwhalen> good.. alright, lets move on.. 16:30:16 <pwhalen> #topic 3) ==== U-Boot Status ==== 16:30:44 <pwhalen> #info Latest - uboot-tools-2016.03-0.4.rc3.fc24 16:32:10 <pwhalen> generally looking sane overall. Ive hit an issue with the Banana Pi, data aborts after loading the dtb. this has happened previously with the debug enabled kernels. 16:32:17 <pbrobinson> basically test and report issues 16:34:09 <pwhalen> #topic 4) == F24 testing == 16:34:21 <pwhalen> #link https://fedoraproject.org/wiki/Test_Results:Fedora_24_Branched_20160308_Summary?rd=Test_Results:Current_Summary 16:34:55 <pwhalen> todays compose has been nominated (armhfp), please test and add results to the wiki. 16:35:41 <pwhalen> #info Please test and add results to the wiki. 16:35:46 <pbrobinson> we should be able to add aarch64 to that later in the week 16:36:07 <pwhalen> very nice, let me know and ill set up some pages for us 16:36:34 <pwhalen> #topic 5) == Open Floor == 16:36:52 <hrw> pwhalen: how we are with m400 builders? 16:36:53 <pwhalen> thats all i had for today, anything else worth mentioning ? 16:37:11 <pbrobinson> hrw: won't be touched likely until post alpha 16:37:24 <pbrobinson> hrw: I have a bunch of things higher up the priority list 16:37:33 <pbrobinson> pwhalen: nothing from me 16:37:36 <hrw> pbrobinson: ok, but are they online at all? 16:37:48 <pbrobinson> hrw: define online 16:38:05 <hrw> ssh fedora-m400-d34.somewhere 16:38:30 <hrw> if not used than it would be faster to build in 32GB tmpfs than on rotating hdd 16:38:36 <pbrobinson> hrw: nope, they're on one of the build networks which is restricted 16:38:46 <hrw> ok 16:38:57 <hrw> still was worth asking ;D 16:39:16 <hrw> nothing more from me 16:39:45 <pwhalen> ok.. last call.... 16:40:27 <pwhalen> #endmeeting