15:00:19 <pwhalen> #startmeeting Fedora ARM & AArch64 Status Meeting
15:00:19 <zodbot> Meeting started Tue Aug 18 15:00:19 2015 UTC.  The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:19 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:19 <pwhalen> #chair pwhalen bconoboy pbrobinson dgilmore hrw jsmith kyle dmarlin
15:00:19 <zodbot> Current chairs: bconoboy dgilmore dmarlin hrw jsmith kyle pbrobinson pwhalen
15:00:31 * bconoboy waves
15:00:33 <pwhalen> good morning! whos here today?
15:00:40 <zsun> .hellomynameis zsun
15:00:40 <zodbot> zsun: zsun 'Ziqian SUN (Zamir)' <sztsian@gmail.com>
15:00:52 * zsun good evening
15:01:02 * pbrobinson waves
15:01:10 <ajax> here, but only to troll about a specific bug
15:02:07 <pbrobinson> ajax: you know you don't have to wait for meetings to do that right?
15:02:18 <ajax> sure, but i saw the meeting announced ;)
15:02:36 <pbrobinson> :)
15:03:20 <pwhalen> alright, lets get started.
15:03:22 <pwhalen> #topic 1) ==== Package Status & Issues  ====
15:03:49 <pwhalen> pbrobinson, in the 5 minutes since you walked in the door.. hows it all looking? :)
15:03:56 <pbrobinson> nothing major here to report, still haven't had time to do eclipse bootstrap
15:04:11 <pwhalen> our resident 'fixer' is wanderign the streets of Toronto right now (hrw)
15:04:14 <pbrobinson> plan on trying to get the golang bootstrap done this week too
15:04:38 <pbrobinson> that's it that I'm aware of
15:04:53 <pbrobinson> we're about 500 packages behind
15:05:17 <pwhalen> #info pbrobinson to tackle the golang/eclipse bootstraps this week
15:05:19 <pbrobinson> sorry, 250 behind, 500 miissing
15:06:27 <pwhalen> anyone aware of any others?
15:06:56 <pwhalen> ok, moving on...
15:06:58 <pwhalen> #topic 2) ==== Kernel Status ====
15:07:19 <pwhalen> #info - FIXED! OOPS - kernel BUG at arch/arm/mm/highmem.c:114!
15:07:35 <dgilmore> hi
15:07:35 <zsun> still can not boot from mSD card on Beagle Bone Black (Alpha)
15:07:43 <pwhalen> #link https://bugzilla.redhat.com/show_bug.cgi?id=1250613
15:07:43 <pbrobinson> yes, a big shout out to msalter for finding that one
15:07:53 <pwhalen> morning dgilmore
15:08:20 <dgilmore> pwhalen: we still need to get that kernel stable
15:08:29 <dgilmore> with the highmem fix
15:08:29 <pbrobinson> zsun, OK, I've not had time to look, we need a nightly image with the above fix so I can better debug but I've been traveling the last 10 days
15:08:34 <pwhalen> good point
15:09:01 <zsun> pbrobinson: okay. I know there are always a lot of meetings in August ;)
15:09:03 <pbrobinson> dgilmore: yep, I know, basically what I meant about image with it in
15:09:47 * jsmith is tied up in a ${DAYJOB} meeting and can't attend today
15:10:05 <pbrobinson> jsmith: no issues
15:10:52 <pbrobinson> anyone else have kernel issues, queries, have done testing on any particular device to report good/bad?
15:12:18 <pwhalen> #info kernel-4.2.0-0.rc7.git0.1.fc23 is out, please test, provide feedback - https://koji.fedoraproject.org/koji/buildinfo?buildID=677859
15:12:48 <pwhalen> quick look yesterday I didnt see any regression, but need to test across the board.
15:13:33 <zsun> ok, will test it my tomorrow
15:13:37 <pbrobinson> now I'm home I'll be testing it widely this week
15:13:43 <pwhalen> thanks zsun!
15:13:55 <pwhalen> #topic 3) ==== F23 - Alpha RC1 for AArch64 ====
15:14:05 <pwhalen> #info Testing so far - https://fedoraproject.org/wiki/Architectures/AArch64/QA/23_Alpha_RC1
15:14:23 <pbrobinson> so I believe we currently have a single issue with PXE boot on Seattle
15:14:35 <pwhalen> so far, dmarlin found one bug - https://bugzilla.redhat.com/show_bug.cgi?id=1251600
15:14:37 <pwhalen> right
15:15:31 <pwhalen> this affects only seattle pxe boots, mustang pxe'd fine. seattle installed from DVD ok
15:15:44 <pwhalen> with an easy work around for pxe, i dont see it as a blocker.
15:16:10 <pbrobinson> me either, anyone else with thoughts?
15:17:16 <pwhalen> i also dont think most people rotate those bits, specifically - BOOTAA64.EFI
15:19:09 <pwhalen> ok, please keep up the testing. remember, you can test without hardware - https://fedoraproject.org/wiki/Architectures/AArch64/F23/Installation#Install_with_QEMU
15:19:42 <pwhalen> #info Testing F23 Alpha RC1 without Aarch64 hardware - https://fedoraproject.org/wiki/Architectures/AArch64/F23/Installation#Install_with_QEMU
15:20:35 <pwhalen> #topic 4) == Open Floor ==
15:21:02 <pwhalen> thats all I had for today, reminder to send me agenda items prior to the meeting..
15:21:07 <pbrobinson> ajax: you had a issue?
15:21:08 <ajax> 1239574 - icu ftbfs in f23
15:21:28 <ajax> which is falling down on arm for no particularly obvious reason
15:21:29 <pbrobinson> OK, anything in particular?
15:21:44 <ajax> erack could use some guidance getting to an arm machine to test i think
15:22:03 <pbrobinson> ajax: pwhalen can get that sorted for you
15:22:16 <bconoboy> Anybody want to recap arm related flock activities?
15:22:28 <ajax> i only noticed because a) it's blocking some cve updates b) it's also not hardened as a result
15:22:33 <pbrobinson> bconoboy: lots!
15:23:17 <bconoboy> pbrobinson: OK, go for it- or I can say stuff
15:23:19 <pbrobinson> ajax: pwhalen can sort you access to a machine
15:23:31 <pbrobinson> bconoboy: please do
15:23:42 <bconoboy> OK, so Flock was last week, ended Saturday
15:24:10 <bconoboy> pbrobinson did a state of the architecture talk for POWER/ARM, should be available on youtube when the rest of the videos go up
15:24:47 <bconoboy> One topic that came up both at this talk, and at a "meet fesco" was the status of secondary architectures in koji
15:25:18 <pbrobinson> dgilmore did a talk on u-boot too where jonmasters admitted he was wrong.... that was a highlight for me at least ;-)
15:25:27 <bconoboy> :-)
15:25:40 <bconoboy> There is an idea brewing that primary/secondary status should be refactored into the fedora.next methodology
15:26:07 <bconoboy> So instead of architectures being primary/secondary, spins (verbiage?) should be primary/secondary
15:26:21 <pbrobinson> bconoboy: and also an adjustment of the idea that koji instances are part of that divider
15:26:38 <bconoboy> What follows from that idea is that koji is the wrong place to assert primary/secondary status
15:27:14 <bconoboy> Which means moving aarch64, ppc64, ppc64le and possibly s390x into the main koji server
15:27:43 <bconoboy> From there it would be up to us to support which spins we think armhfp and aarch64 are suitable
15:27:57 <bconoboy> This makes a *lot* of sense to me
15:28:03 <pwhalen> +1
15:28:31 <bconoboy> We haven't dug into the details too far, but I think the status of primary vs secondary then becomes about whether the health of a spin blocks release
15:29:00 <bconoboy> So aarch64 could be primary on server, armhfp could be primary on xfce, that sort of thing.
15:29:00 <pbrobinson> bconoboy: I think you means more Edition as opposed to spin
15:29:10 <bconoboy> Edition, that's the word (Still recovering:-)
15:29:45 <bconoboy> People at the arm/ppc session didn't seem to mind this idea at all, but why would they?  The interesting thing is that people at the Fesco meeting seemed to think it was a good idea too.
15:30:19 <bconoboy> So I think we want to move forward on this, and soon
15:30:34 <dgilmore> Server WG is talking about essentially the same thing right now :)
15:30:42 <bconoboy> dgilmore: excellent
15:31:12 <bconoboy> It just makes a lot of sense- the primary/secondary thing is really a primitive form of fedora.next, it's time to fedora.next all the architectures.
15:31:23 <bconoboy> pbrobinson: Anything to add? Are you point on this? :-)
15:33:01 <bconoboy> OK, that's it for now
15:33:02 <ajax> i wasn't there for flock, but in a fesco sense, i'm all for it too
15:33:14 <ajax> the secondary builders have always been confusing and awkward
15:33:20 <bconoboy> Also, I think jcm saying "I was wrong" will be on dgilmore's uboot talk, so look for a link from that soon ;-)
15:33:38 <bconoboy> ajax: woot
15:33:41 <dgilmore> bconoboy: indeed, it was a unexpected thing that did bring me joy
15:34:02 <bconoboy> dgilmore: If the youtube video picks it up clearly it needs the autotune treatment
15:34:15 <dgilmore> bconoboy: :) indeed
15:34:25 <bconoboy> pwhalen: back to you?
15:34:40 <pwhalen> all i had, does anyone have anything else?
15:36:58 <pwhalen> going once..
15:37:47 <pwhalen> #endmeeting