20:00:17 <pwhalen> #startmeeting Fedora ARM weekly status meeting
20:00:17 <zodbot> Meeting started Wed Apr 10 20:00:17 2013 UTC.  The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:17 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:17 <pwhalen> #chair pwhalen jonmasters bconoboy ctyler pbrobinson dgilmore
20:00:17 <zodbot> Current chairs: bconoboy ctyler dgilmore jonmasters pbrobinson pwhalen
20:00:27 <pwhalen> .fas pwhalen
20:00:29 <zodbot> pwhalen: pwhalen 'Paul Whalen' <pwhalen@redhat.com>
20:00:40 <pwhalen> good afternoon, who's around today?
20:00:45 * masta looks in
20:00:48 * masta waves
20:01:00 <masta> .fas parasense
20:01:01 <zodbot> masta: parasense 'Jon Disnard' <jdisnard@gmail.com>
20:01:23 <masta> dmarlin is on PTO today and might not join
20:01:41 <pwhalen> jonmasters mentioned he would be joining a few minutes late today
20:02:19 <jcapik> .fas jcapik
20:02:20 <zodbot> jcapik: jcapik 'Jaromír Cápík' <jcapik@redhat.com>
20:02:25 <pwhalen> alright, as people filter in....
20:02:34 <pwhalen> #topic 0) Status of ACTION items from our previous meeting
20:02:48 <pwhalen> #info pwhalen to work with pbrobinson to publish weekly problem package reports
20:02:48 <pwhalen> #info INPROGRESS - list composed, narrowing down list of most relevant packages.
20:03:19 <pwhalen> #info masta, j_dulaney, oatley to try upstream kernel on a10 devices
20:03:24 <pbrobinson> I have the list I'm just working with pwhalen to categorise them and group them into deps
20:03:53 <pwhalen> oatley, j_dulaney, masta , I wasn't able to sync with any of you, how did it go?
20:04:24 <dgilmore> hi
20:04:33 <pbrobinson> masta, j_dulaney, oatley: the core SoC sunxi options are enabled but it's likely that we need more drivers enabled
20:04:58 * whenry is lurking
20:05:02 <masta> I have not had the time to investigate the sunxi boards, but I still do plan to do that
20:05:21 <pwhalen> ok, lets revisit next week
20:05:35 <pwhalen> #action masta, j_dulaney, oatley to try upstream kernel on a10 devices
20:05:39 <masta> I've got gooseberry, and mk802... most like will use gooseberry as it hase uart connector and things like that
20:06:12 <pbrobinson> masta: once we have a f19 minimal image I'm going to test on my Cubieboard with it
20:06:15 <pwhalen> #info bconoboy to draft message to devel-announce requesting aarch64 config.guess/sub support go into f19
20:06:15 <pwhalen> #info bconoboy off this week, follow-up next week.
20:06:29 <pwhalen> anything else from last week?
20:06:36 <pbrobinson> yes!
20:07:00 * pbrobinson _FINALLY_ got around to reviewing the vboot-utils package for masta :)
20:07:35 <pwhalen> cool, approved?
20:07:38 <pbrobinson> yep
20:07:48 <pbrobinson> awaiting git creation
20:08:10 <pwhalen> #info pbrobinson reviewed vboot-utils for masta, awaiting git creation
20:08:22 <pwhalen> next?
20:08:29 <pwhalen> #topic 1) Problem packages
20:08:30 <pbrobinson> problem packages
20:08:39 <jonmasters_> .fas jonmasters
20:08:40 <zodbot> jonmasters_: jcm 'Jon Masters' <jonathan@jonmasters.org>
20:08:44 <pbrobinson> so we're looking pretty god damn good with the packages
20:09:04 <pbrobinson> tog-pegasus is out main blocker to date
20:09:15 <pbrobinson> all ruby packages are closed out
20:09:16 <masta> pwhalen: I think there was something about testing omap kernels on panda?
20:09:27 <pwhalen> #info looking pretty darn good with the packages
20:09:51 <pwhalen> #info tog-pegasus continues to block
20:10:07 <pbrobinson> we've got some new fixes in testing awaiting in the wings
20:10:15 <nirik> would it help to get a strace of the block on tog-pegasus?
20:10:22 * nirik arrives late to the party
20:10:26 <pwhalen> masta, sorry?
20:10:44 <pbrobinson> masta: lets deal with kernel later
20:10:57 <masta> ok
20:11:01 <masta> acknowledged
20:11:04 <masta> my bad
20:11:54 <masta> nirik: what do you mean?
20:11:55 <ahs3> .fas ahs3
20:11:55 <pwhalen> not sure if dmarlin is around right now but he's been looking into tog-pegasus
20:11:55 <zodbot> ahs3: ahs3 'Al Stone' <ahs3@redhat.com>
20:12:01 <pbrobinson> and I don't have anything else, would love some more help with tog
20:12:23 <nirik> masta: well, tog-pegasus is hanging in tests... I could get a strace of the process from the builder... might help to narrow down whats looping.
20:12:28 * pbrobinson waves to ahs3 and jonmasters
20:12:32 <nirik> is there a bug to attach that to?
20:13:15 * nirik guesses the last one was canceled.
20:13:45 <pwhalen> anyone have it handy?
20:13:51 * pbrobinson is looking
20:13:53 <masta> nirik: oh yes that might be helpfull... I'd just build in mock and do the make tests
20:14:04 <pbrobinson> https://bugzilla.redhat.com/show_bug.cgi?id=922770
20:14:14 <nirik> masta: sure if you can duplicate it there...
20:15:00 <jonmasters> .fas jonmasters
20:15:01 <zodbot> jonmasters: jcm 'Jon Masters' <jonathan@jonmasters.org>
20:15:02 <jonmasters> back, sorry
20:15:03 <pwhalen> #link tog-pegasus BZ - https://bugzilla.redhat.com/show_bug.cgi?id=922770
20:15:04 * jonmasters gave dmarlin some pointers about atomics in pegasus
20:15:06 <jonmasters> but there's an unrelated issue that still needs love
20:15:08 <jonmasters> nirik: dmarlin has a pointer to a BZ
20:15:10 <jonmasters> nirik: sure, capture the trace, but I know David has also done that
20:15:12 <jonmasters> (he has a local system in HSV that he is running tests on)
20:15:14 <jonmasters> btw, is my mic on? I just reset my bot - am I here?
20:15:30 <jonmasters> cool, thanks pwhalen
20:15:39 <nirik> cool. If there's anything further I can provide let me know. Sounds like many eyes are already on it.
20:16:07 <pwhalen> ok, think that summed it up, any other problem packages/
20:16:07 <pbrobinson> jonmasters: yes you're here
20:16:09 <kalev> pbrobinson: everything looking good on arm with GNOME-related packages?
20:16:12 <jonmasters> pbrobinson: thanks
20:16:31 <pbrobinson> kalev: yes, I think so, they've built at least
20:16:35 <kalev> excellent
20:16:50 * ahs3 refuses to deal with existential questions this early in the day
20:17:08 <pbrobinson> ahs3: what timezone are you in?
20:17:22 <ahs3> pbrobinson: GMT-7 :)
20:17:32 <pwhalen> #topic 2) Aarch64 patching status - minimal install package set
20:17:59 <pbrobinson> by minimal do you mean buildroot + minimal install?
20:18:33 <pwhalen> believe buildroot, jonmasters ?
20:18:52 <jonmasters> sec
20:19:08 <jonmasters> minimal install
20:19:42 <jonmasters> pbrobinson: this is the thing we discussed last week, really just an opportunity to ask dgilmore if he's had chance to ponder what the minimal number of remaining packages for F19 looks like
20:19:48 <pbrobinson> so do we have a package list for that and can we cross reference it with the list of open bugs against the blocker to work out what is remaining?
20:20:20 <jonmasters> pbrobinson: Dennis has those things and was going to look at making such a list
20:20:27 <dgilmore> jonmasters: ive not yet
20:20:57 <jonmasters> dgilmore: ok, not a problem. Let's move on and sync on this one later then?
20:21:13 <pbrobinson> surely dgilmore isn't the only one that can produce that list and I suspect he's fairly loaded up with mainline and ARM alpha
20:21:20 <pbrobinson> so can someone else pick this up?
20:22:00 <pbrobinson> jonmasters: any one in your team not fully utilised at the moment?
20:22:09 * jonmasters suggests we put Brendan on the hook as he's not here to defend himself
20:22:10 <pbrobinson> see my last comment :)
20:22:15 <pbrobinson> haha
20:22:19 <pbrobinson> works for me :)
20:22:39 <jonmasters> #action bconoboy to work out remaining minimal F19 package set requiring aarch64 autotools patching
20:22:42 <jonmasters> :P
20:23:04 <pwhalen> #topic 3) Testing F19 Alpha install tree candidate
20:23:11 <pwhalen> #link http://armpkgs.fedoraproject.org/mash/stage/19-Alpha-RC2/Fedora/
20:23:28 <dgilmore> the install tree matches primary
20:23:43 <dgilmore> test and see what works and doesnt
20:23:51 <pwhalen> dgilmore composed a new install tree that is a candidate for our alpha. I plan on doing an installation on highbank.. currently we have no images
20:24:27 <dgilmore> image composition is broken somewhere in anaconda
20:24:43 <dgilmore> it could be livemedia-creator where its broken
20:25:23 <pbrobinson> OK, so that's pretty good progress. Do we have someone that's looking at it?
20:26:01 <dgilmore> i was talking with the anaconda guys earlier but they didnt provide any help on how to debug it further
20:26:11 <pbrobinson> useful!
20:26:36 <dgilmore> they looked at all teh logs and said it all looks normal
20:27:34 <pwhalen> I'll report back with results from testing on the highbank..
20:27:43 <pwhalen> next?
20:27:53 <pbrobinson> what's next on the agenda?
20:27:57 <pwhalen> #topic 4) Kernel status update
20:28:07 <pbrobinson> so 3.9 is slowly shaping up
20:28:15 <dgilmore> pwhalen: sure, install should work on trimslice also to a point
20:28:22 <pbrobinson> less than a week ago it only booted on vexpress
20:28:35 <pbrobinson> but as of yesterday it's booting on tegra and highbank
20:28:41 <pbrobinson> but it's bloody ropey
20:28:50 <pbrobinson> with a number of kernel oops
20:28:56 <pbrobinson> on both of those platforms
20:29:04 <pbrobinson> next to focus is omap
20:29:30 <pbrobinson> once we've got omap working I'll be focusing on mvebu/am35xx (bbone)
20:29:54 <pbrobinson> so we're getting there, not as far forward as I would like but we're at least moving forwards :)
20:30:08 <pbrobinson> on the 3.8.x front I think it's looking very good
20:30:08 <masta> dgilmore: from the anaconda storage logs , they do not look normal, there is an error when applying the disklabel
20:30:22 <jonmasters> pwhalen mentioned he had done some omap tests, but I believe it was hanging at boot
20:30:26 <masta> sry, responding to previous topic
20:30:28 <dgilmore> masta: the anaconda guys said it looked normal
20:30:57 * jonmasters knows Peter needs him to look at it this week too, haven't forgotten
20:31:00 <pbrobinson> I'm not aware of any outstanding 3.8 issues so I'm now basically ignoring that
20:31:18 <pbrobinson> jonmasters: I think I'll really need your help with omap.
20:31:19 <pwhalen> kernel-3.9.0-0.rc6.git0.1.fc20 tested on tegra and highbank so far - highbank booted with network, an attempt to fpaste dmesg and it crashed
20:31:36 <jonmasters> pbrobinson: yea, I got that from Paul earlier too in his internal status summary, ok
20:31:39 <pbrobinson> jonmasters: my pandaES has problems with anything > 3.6
20:31:43 <jonmasters> indeed
20:31:51 <dgilmore> im going to test on my panda
20:31:54 <jonmasters> pwhalen: you were going to play with dtbs some more...any dice?
20:31:57 <pbrobinson> so I'd like to try and work out how I can make it work
20:32:29 <pwhalen> on the panda(es) ** Unable to read "omap4-panda-es.dtb" from mmc 0:1 **
20:33:01 <pwhalen> still looking at it, was just before the meeting and wanted to test the highbank before we started
20:33:08 <pbrobinson> pwhalen: dgilmore just mentioned we might not have dtb support in the panda uboot's
20:33:35 <pwhalen> previously its loaded it before, and just not gone further then Starting the kernel
20:33:37 <dgilmore> pbrobinson: no that was the armada-xp system
20:33:39 <masta> I think the panda boards wanted to have appended dtb
20:33:42 <pbrobinson> ah
20:33:43 * masta hides
20:33:47 <pbrobinson> sorry, misinterpret
20:35:00 <masta> In the past I've got the DTB to load fine via load addr, but the /proc/device-tree (or whatever) did not exist, but it did with an appended
20:35:11 <pbrobinson> weird
20:35:24 <pwhalen> jonmasters, fpaste of highbank crash http://paste.fedoraproject.org/7125/36562301/
20:35:30 <dgilmore> masta: ive personally only gotten dtb to work on panda by appending
20:35:34 <jonmasters> masta: note also the /proc/device-tree stuff has been rewritten
20:35:42 <masta> the above was on omap, fyi
20:35:48 <jonmasters> if anyone's interested, Grant switched device tree over to use sysfs
20:36:01 <jonmasters> so /proc/device-tree is now a link to sysfs going forward, just FYI
20:36:17 <pbrobinson> jonmasters: how does that work, so nice it was widely advertised
20:36:23 <masta> whoa, so what is this sysfs ?
20:36:41 <dgilmore> jonmasters: i dont care how its implemented as long as it works
20:36:46 <masta> moves from /proc/device-tree (or whatever) to /sys/device-tree ?
20:36:47 <dgilmore> and is stable
20:36:56 <jonmasters> pbrobinson: well, sysfs contents use the standard kernel object model, so it's a nice thing to switch and apparently transparent because /proc/device-tree will still work
20:37:18 <jonmasters> but I'm encouraged to see that cleanup...anyway, we can look at the append vs. non-append issue
20:37:30 <masta> ok good, the /proc-device tree was my metric to test working dtb
20:37:35 <jonmasters> pbrobinson: yea, I only saw that because I was reading through all old LKML mail at the weekend!
20:37:48 <jonmasters> pwhalen: looking at trace
20:39:37 <pbrobinson> right so what else do we have to cover in the meeting?
20:39:41 <pwhalen> shall we move on ?
20:39:48 <pwhalen> open floor was next
20:40:02 <pwhalen> #topic 5) Your topic here
20:40:31 <jonmasters> pwhalen: that backtrace above - the box doesn't hang though, right? It's still alive after?
20:40:50 <masta> dgilmore: who in anaconda team did you work with regarding image creating and fine looking logs?
20:41:02 <pwhalen> jonmasters, its dead, haven't rebooted
20:41:07 <dgilmore> masta: dlehman
20:41:19 <masta> dgilmore: thx
20:41:36 <pbrobinson> any other business or should we close the meeting and head on over to #fedora-arm?
20:41:38 <jonmasters> pwhalen: oh ok, can you leave that node for a bit? I might be able to poke, won't be until late tonight. Probably need Calxeda to look. It's stalling in a DMA transfer from the network
20:41:45 <jonmasters> sorry, I'm on a tangent :)
20:41:48 <jonmasters> sorry Peter!
20:42:05 <masta> I'm good to end early
20:42:13 <pbrobinson> jonmasters: let's close this out and move on out
20:42:40 <pwhalen> sounds good.
20:42:41 <pwhalen> #endmeeting