#title #ubuntu-meeting: ubuntu-arm

Meeting started by ogra_ at 15:08:02 UTC.  The full logs are available
at
http://ubottu.com/meetingology/logs/ubuntu-meeting/2011/ubuntu-meeting.2011-12-08-15.08.log.html
.



== Meeting summary ==

 *spec status, http://status.ubuntu.com/ubuntu-precise/ubuntu-arm.html

 *ARM Server Status (NCommander, Daviey)

 *Kernel Status (cooloney, ppisati)

 *ARM Porting/FTBFS status (NCommander, janimo)

 *ARM Hardfloat status (infinity)

 *ARM Image Status (ogra, NCommander)

 *QA Status (GrueMaster, mahmoh)

 *Linaro Updates (rsalveti)
''ACTION:'' rsalveti to track the gles driver status for armhf for the vendors we support  (rsalveti, 15:45:54)
''ACTION:'' work with jcrigby to update the mx53 kernel  (rsalveti, 15:46:40)
''ACTION:'' rsalveti to kick someone to fix bug 838200  (GrueMaster, 15:46:52)

 *AOB



Meeting ended at 15:48:09 UTC.



== Votes ==




== Action items ==

 * rsalveti to track the gles driver status for armhf for the vendors we support
 * work with jcrigby to update the mx53 kernel
 * rsalveti to kick someone to fix bug 838200



== Action items, by person ==

 * rsalveti
 ** rsalveti to track the gles driver status for armhf for the vendors we support
 ** rsalveti to kick someone to fix bug 838200



== People present (lines said) ==

 * ogra_ (128)
 * infinity (42)
 * rsalveti (37)
 * GrueMaster (31)
 * janimo (13)
 * ppisati (12)
 * meetingology (6)
 * doko (2)
 * ahs3 (1)
 * ubottu (1)



== Full Log ==


 15:08:02 <ogra_> #startmeeting ubuntu-arm

 15:08:02 <meetingology> Meeting started Thu Dec  8 15:08:02 2011 UTC.  The chair is ogra_. Information about MeetBot at http://wiki.ubuntu.com/AlanBell/mootbot.

 15:08:02 <meetingology> 

 15:08:02 <meetingology> Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired

 15:08:20 <ogra_> agenda is at https://wiki.ubuntu.com/ARM/Meeting/2011/20111208

 15:08:31 <ogra_> seems ther are no action items

 15:08:35 <ogra_> :)

 15:08:48 <ogra_> so lets go over the standing items ...

 15:09:18 <ogra_> [topic] spec status, http://status.ubuntu.com/ubuntu-precise/ubuntu-arm.html

 15:09:45 <ogra_> doesnt look bad ... are all specs on there now ?

 15:09:51 <ogra_> shout if you are missing any

 15:09:56 <infinity> I should probably fill in some work items on my specs before I get yelled at, shouldn't I?

 15:10:04 <ogra_> might help

 15:10:12 <infinity> And I should break the livefs-in-soyuz WI out of image-pipeline into a spec. :/

 15:10:23 <infinity> Cause, well, a month of work isn't a WI.

 15:10:24 <ogra_> also make sure they are approved, have a prio and are assigned to precise by davidm

 15:10:54 <GrueMaster> My bare-metal spec is still under review.

 15:11:08 <ogra_> poke davidm please, once he is back

 15:11:24 <GrueMaster> Not that it matters much.  It was mainly designed to be informational.

 15:11:30 <ogra_> i really would like to have the tracker working before i start my vacation in a week

 15:11:32 <infinity> ogra_: I'm on holidays in a week and a day.  Can I count on you to bug me to make sure my specs are sane before that?

 15:11:48 <infinity> (Today's a bit too busy on the armhf porting front for me to do too much paperwork/drafting)

 15:11:54 <ogra_> infinity, will try to, i'm on vac. from 19th on myself

 15:12:07 <ogra_> moving on then ...

 15:12:07 <infinity> ogra_: From the 16th here. :)

 15:12:19 <ogra_> well, thats the same :P

 15:12:27 <infinity> ogra_: No, as in, I'm taking the 16th off.

 15:12:39 <ogra_> [topic] ARM Server Status (NCommander, Daviey)

 15:12:47 <ogra_> aha ... yeah, ok, thats different

 15:13:11 <ogra_> Daviey, any server news (since NCommander seems to have vanished off the face of the earth)

 15:13:43 * ogra_ suspects not ...

 15:14:04 <ogra_> well, lets move on ... we can come back to it if someone serverish shows up

 15:14:12 <ogra_> [topic] Kernel Status (cooloney, ppisati)

 15:14:28 <ppisati> armhf support has hit the archive (as you may have noticed)

 15:14:35 <ppisati> and 3.2.0-x too

 15:14:38 <ogra_> apart from mx5 i think

 15:14:39 <infinity> Has anyone taken the 3.2 omap4 for a solid spin on a Panda yet?

 15:14:44 * ahs3 is server-ish but has no news

 15:15:00 <ppisati> infinity: i found a bug this morning

 15:15:10 <ppisati> ogra_: mx5? is it community supported or what?

 15:15:16 <ogra_> ppisati, linaro

 15:15:17 <infinity> Now that mine's not stuck building ghc, I can make it grumpy with 3.2 and armhf.

 15:15:33 <infinity> ppisati: Yay bugs.  Is it any worse than the ones we currently have? :)

 15:15:44 <ppisati> infinity: panic on reboot if a usbhdd is used, i already pushed a new kernel with an ext4 fix and that too

 15:16:07 <ppisati> if pulling the cord on reboot is not a problem, feel free to update now

 15:16:09 <ogra_> we only provide SD images anyway ... just ignore it :P

 15:16:14 <infinity> ppisati: Snazzy.  I need to test if 3.2.x fixes all my NPTL/TLS woes, and I suspect it will, from what markos has been saying.

 15:16:15 <ppisati> else, wait until tomorrow

 15:16:21 <ppisati> ok

 15:16:27 <GrueMaster> No, don't ignore it!

 15:16:31 <ogra_> heh

 15:16:39 * ogra_ wasnt serious

 15:16:47 <ppisati> ogra_: do you want me to do the armhf stuff for mx5?

 15:16:51 <janimo> sorry fior being late, hhello

 15:16:57 <ogra_> ppisati, ask jcrigby

 15:17:00 <ppisati> ok

 15:17:02 <infinity> ppisati: Should be jcrigby's alley, no?

 15:17:02 <ogra_> janimo, np

 15:17:05 <janimo> ppisati, is that not linaro task?

 15:17:19 <ogra_> janimo, i guess john appreciated help

 15:17:23 <janimo> zalong with a 3.0 kernel for mx5 whic I hear is available

 15:17:27 * ppisati leaves the mx5 to its destiny...

 15:17:28 <infinity> Anyhow, we're in no huge rush on mx5/armhf.  We have Panda and ac100 to play with.

 15:17:28 <GrueMaster> I would assume we want armHF on mx5, since everything else is moving off armel.

 15:17:28 <ppisati> :)

 15:17:31 <ogra_> and we would appreciate being able to build images :)

 15:17:32 <infinity> And lots of bugs to shake out.

 15:17:43 <ogra_> yeah

 15:18:09 <ogra_> GrueMaster, we do, but we're suddenly so ahead of time that it doesnt bite us to wait another week or two

 15:18:24 <infinity> Well, and we have plenty of userspace bugs to shake out.

 15:18:28 <ogra_> especially for an image thats a "tech preview"

 15:18:35 <infinity> Kernel space is identical from armel to armhf, so I'm less concerned.

 15:18:41 <ogra_> any other kernel bits ?

 15:18:44 <infinity> We'll get there.

 15:18:46 <ogra_> else i'll move

 15:19:03 <infinity> ogra_: Boogie on.

 15:19:05 <ogra_> [topic] ARM Porting/FTBFS status (NCommander, janimo)

 15:19:12 <ogra_> (and everyone else :) )

 15:19:20 <infinity> Thanks to armhf, the armel FTBFS list is looking better too. :P

 15:19:23 <infinity> In fact, the whole distro is.

 15:19:24 <ogra_> armhf ftbfs looks very very good

 15:19:30 <infinity> armhf == free rebuild test.

 15:19:33 <janimo> gave back a few things, next up libo test, once I install armhf on the board

 15:19:34 <ogra_> yep

 15:19:58 <infinity> libreoffice needs love.  Not sure if doko and Sweetshark are still going back and forth on that.

 15:20:04 <infinity> I hope so.

 15:20:09 * ogra_ looked into some bits and wasted a lot of time for getting an ubiquity build env up today...

 15:20:10 <infinity> I'm working on ghc and gnat.

 15:20:20 <janimo> infinity, I'll look at it, doko said he's done as far as he's concerned

 15:20:23 <infinity> And we have people pondering looking at fpc.

 15:20:51 <ogra_> anything else ?

 15:21:04 <GrueMaster> netinstall.

 15:21:09 <infinity> (fpc needs a full port and bootstrap at the upstream level, if you speak fluent pascal and arm ASM without context-switching between the two, speak up)

 15:21:12 <ogra_> ftbfs ??

 15:21:38 <GrueMaster> No, fail to install.  libc6 udeb.

 15:21:44 * ogra_ moves on

 15:21:49 <ogra_> [topic] ARM Hardfloat status (infinity)

 15:21:53 <ogra_> :)

 15:22:03 <ogra_> i guess there isnt much to add

 15:22:03 <infinity> There we go.  Now you can bring up netinst. ;)

 15:22:12 <ogra_> wouldnt that belong into images ?

 15:22:25 <infinity> other than "things are going really, really well with armhf" (much better than expected), we have some bugs to shake out.  Installer oopses, etc.

 15:22:31 <GrueMaster> Someone is moving the meeting too fast for me to type.

 15:22:42 <infinity> ogra_: It's both?  I dunno. :P

 15:22:48 <infinity> ogra_: It's HF-specific, I believe.

 15:22:52 <ogra_> the queue seems to have ~3000 packages ... is that accurate ?

 15:22:56 <infinity> ogra_: Yeah.

 15:22:59 <ogra_> wow

 15:23:01 <doko> qtmobility needs fixing

 15:23:05 <ogra_> so we should be doen by the weekend

 15:23:19 <infinity> ogra_: FSVO "done"... There's lots to do still.

 15:23:27 <ogra_> doko, universe *shrug* .... that has time

 15:23:36 <doko> ogra_, kubuntu package set

 15:23:42 <infinity> Universe is big.  Fix early, fix often.

 15:23:44 <ogra_> argh

 15:24:07 <ogra_> how about the kubuntu guys ? did someone talk to them ?

 15:24:23 <infinity> It just has a bad symbols file.

 15:24:26 <infinity> 5 minute fix.

 15:24:27 <infinity> Move on.

 15:24:30 <ogra_> k

 15:24:41 <ogra_> [topic] ARM Image Status (ogra, NCommander)

 15:24:49 <ogra_> started hf testbuilds

 15:25:00 <ogra_> which was somehow trashed in the middle by broken packages

 15:25:09 <ogra_> a new omap4 build is just running now

 15:25:27 <ogra_> if that succeeds too (ac100 works already) i will switch on autobuilds tomorrow

 15:25:32 <infinity> netinst needs (at least) my mklibs fix uploaded and a d-i rebuild.

 15:25:34 <infinity> But might need more.

 15:25:35 <ogra_> for everything but mx5

 15:25:47 <GrueMaster> yep.   hunspell-en-us broke the world (not just arm)

 15:25:51 <ogra_> yep

 15:25:53 <infinity> It sure did.

 15:25:59 <ogra_> but its fixed

 15:26:03 <ogra_> as is ubiquity

 15:26:25 <ogra_> so i dont expect the currently running omap4 build to fall over

 15:26:53 <ogra_> but we'll see :)

 15:27:01 <ogra_> anything else for images

 15:27:18 * ogra_ waits for GrueMaster to not speeed to fast for the old man :)

 15:27:29 <GrueMaster> Hey!

 15:27:36 <ogra_> *g*

 15:27:46 <ogra_> ok, moving then

 15:27:51 <infinity> We're all old. :P

 15:27:58 <ogra_> [topic] QA Status (GrueMaster, mahmoh)

 15:28:04 <ogra_> GrueMaster, stage is all yours

 15:28:25 <GrueMaster> Still working on automation for SRU testing.  Seems that everytime I get one test working for QRT, another breaks, due to x86 requirements.

 15:28:56 <ogra_> how about your jenkins stuff, when is that due to move to the official jenkins instance ?

 15:29:10 <GrueMaster> And I still don't have a clear scenario on what to test for SRU approval.  Currently I run the QRT kernel tests, and a few other things.

 15:29:28 <ogra_> (doesnt seem to be there yet ... i was looking for it for a report i had to write for david yesterday)

 15:29:36 <GrueMaster> I won't move it to Jenkins until I can get it stabilized here.

 15:29:40 <ogra_> ah, k

 15:29:55 <GrueMaster> And since I have very little visibility with them, it is much harder.

 15:30:00 <ogra_> well SRU approval is indeed tricky

 15:30:11 <ogra_> since it usually refers to one specific bug

 15:30:23 <ogra_> which your test would have to reproduce

 15:30:27 <GrueMaster> Only in that there is no documented "This is what we run to approve".

 15:30:51 <ogra_> well, "does the bug still occur" is what we run i would guess

 15:31:04 <ogra_> no idea if thats even remotely automatable

 15:31:23 <ogra_> you can surely automate regression testing ...

 15:31:25 <GrueMaster> Problem with that is that there are a LOT of tests that were fixed on x86 that still exist on arm.

 15:31:31 <ogra_> but likely not the verifgication

 15:31:56 <GrueMaster> I can only automate what I know (which isn't much).

 15:32:00 <ogra_> right

 15:32:30 <GrueMaster> And when the upstream test scripts keep breaking, I have to manually intervene.

 15:32:36 <GrueMaster> Making automation moot.

 15:32:38 <ogra_> indeed :/

 15:33:13 <GrueMaster> So far, this have been my primary focus since UDS.  And it is taking way too much of my time.

 15:33:47 <ogra_> any chance to get some external help from i.e. linaro ?

 15:34:00 <ogra_> i would assume they suffer similar issues in some areas

 15:34:08 <ogra_> or do they do everything from scratch

 15:34:32 <GrueMaster> It isn't really a linaro thing.  It is mainly working with the security team to fix their wrt test scripts.

 15:34:56 <ogra_> ah, yeah, and thats something linaro doesnt touch at all

 15:35:01 <GrueMaster> And linaro is more interested in performance benchmarking.  Not sure they do any SRU testing.

 15:35:12 <ogra_> they dont do SRUs indeed

 15:35:38 <ogra_> anything else for QA ?

 15:36:08 <GrueMaster> I would like to thank NCommander (if he were here) for providing me a hacked netinstall for Maverick & Natty on Panda.  At least I can automate the install for that part of the SRU testing.

 15:36:31 * ogra_ applauds the absent NCommander

 15:36:43 <GrueMaster> It also has helped expose a bug introduced in another package (procps) that was in proposed.

 15:36:56 <GrueMaster> That affected all platforms.

 15:37:02 <ogra_> great

 15:37:24 <GrueMaster> And I'm done.

 15:37:41 <ogra_> [topic] Linaro Updates (rsalveti)

 15:37:46 <GrueMaster> Nothing more to see here, move along.

 15:37:51 * rsalveti waves

 15:37:57 <ogra_> heys rsalveti

 15:38:08 <rsalveti> guess no much to report during this week

 15:38:18 <rsalveti> marcin was able to push some of the fixes to doko

 15:38:20 <janimo> rsalveti, do you know what the status of mx5 3.0 kernel is?

 15:38:21 <rsalveti> for the cross toolchain

 15:38:44 <rsalveti> janimo: we should be pushing a new 3.1 kernel soon, once jcrigby is done with the ci bp

 15:38:45 <ogra_> rsalveti, with the enormous speed that armhf gets forward we are very likely to swithch asap

 15:38:50 <GrueMaster> Can you nudge your guys to fix the beagleXM network issue?

 15:38:51 <rsalveti> probably in the next few days

 15:38:53 <ogra_> rsalveti, is linaro prepared for that ?

 15:39:07 <rsalveti> ogra_: sure :-)

 15:39:15 <rsalveti> we also want to switch asap :-)

 15:39:17 <ogra_> (wrt binary drivers mainly)

 15:39:18 <janimo> rsalveti, ah great. So 3.1 will be pushed to the archives

 15:39:27 <rsalveti> ogra_: well, that's the complicated part

 15:39:37 <rsalveti> talked with ti but that is still not a priority for them

 15:39:41 <ogra_> rsalveti, i'll bring the panda stuff up in tomorrows call

 15:39:53 <ogra_> but what about beagle ?

 15:39:56 <rsalveti> yeah, would be better to do the follow up tomorrow

 15:40:01 <ogra_> or other vendors

 15:40:13 <rsalveti> ogra_: same thing, they don't care much about armhf atm

 15:40:18 <ogra_> do you have any lever in linaro you could move them with ?

 15:40:23 <ogra_> sigh

 15:40:28 <rsalveti> sure, anmar is our guy for it

 15:40:32 <rsalveti> will ping him about it

 15:40:36 <ogra_> that smells like another release without unity 3d

 15:40:48 <ogra_> even though we have the code now

 15:40:51 <rsalveti> nicolas said he'd try to build the driver for armhf

 15:40:58 <ogra_> nothing we will support can run it

 15:41:09 <rsalveti> and then see how that goes, but if he needs to put some work on that, it'll probably be hard to have it released

 15:41:11 <ogra_> rsalveti, right, as i said, thats for the call rtomorrow

 15:41:12 <rsalveti> yeah

 15:41:26 <janimo> rsalveti, regarding the cross-build of firefox and chromium that riku is doing, do you know what can be done to get the multiarch-fixed libs in precise? They are now in a PPA, and riku has no upload right, but I am not sure if anything is done to push them ATM

 15:41:35 <ogra_> and he is aware that he has all bits for building already

 15:41:57 <rsalveti> janimo: yeah, that's why we have https://blueprints.launchpad.net/linaro-ubuntu/+spec/push-multiarch-changes-for-cross-precise

 15:42:12 <rsalveti> janimo: the goal for this bp is to have at least everything in a bug with the patch attached

 15:42:20 <rsalveti> but atm he's doing well on both ubuntu and debian

 15:42:36 <rsalveti> janimo: and also poked him to update your request for feedback

 15:42:47 <janimo> rsalveti, I saw that but it was not clear to me whether wotk is ongoing or stalled (or not a prio for linaro)

 15:42:53 <janimo> wrt pushing to the main archive

 15:43:05 <rsalveti> janimo: that's the main thing he's working atm

 15:43:08 <janimo> ok

 15:43:11 <rsalveti> he finished the chromium one already

 15:43:17 <janimo> thanks

 15:43:20 <rsalveti> so now he's just working on pushing it to precise

 15:43:26 <rsalveti> feel free to ping him about it anytime

 15:43:42 <rsalveti> guess that's all from my side

 15:43:43 * ogra_ guesses FF has a higher prio on our side

 15:43:52 <rsalveti> so two actions

 15:44:01 <ogra_> at least thats how i understood smagoun

 15:44:05 <janimo> well having all the depends in is what matter a lot too, as others use them

 15:44:14 <rsalveti> yup

 15:44:16 <ogra_> yep

 15:44:44 <ogra_> anything else ? more questions ?

 15:44:49 <rsalveti> nops

 15:44:54 * ogra_ moves to AOB then

 15:44:54 <rsalveti> how to add action with this bot?

 15:45:02 <rsalveti> should use #action?

 15:45:10 <ogra_> i think angled brackets

 15:45:19 <ogra_> but hash might also work

 15:45:31 <ogra_> try it :)

 15:45:42 <infinity> I thought it was square braces? :P

 15:45:54 <rsalveti> #action rsalveti to track the gles driver status for armhf for the vendors we support

 15:45:54 * meetingology rsalveti to track the gles driver status for armhf for the vendors we support

 15:45:56 <ogra_> youre so square :P

 15:45:56 <GrueMaster> Its hash, but has to be done by the moderator.

 15:45:58 <rsalveti> yup :-)

 15:46:04 <GrueMaster> Or that.

 15:46:06 <ogra_> GrueMaster, apparentl not :)

 15:46:21 <ogra_> rsalveti, you said two actions ?

 15:46:37 * ogra_ waits for the second before he moves to AOB

 15:46:40 <rsalveti> #action work with jcrigby to update the mx53 kernel

 15:46:40 * meetingology work with jcrigby to update the mx53 kernel

 15:46:44 <rsalveti> done!

 15:46:45 <ogra_> great :)

 15:46:52 <GrueMaster> #action rsalveti to kick someone to fix bug 838200

 15:46:52 * meetingology rsalveti to kick someone to fix bug 838200

 15:46:53 <ubottu> Launchpad bug 838200 in u-boot-linaro (Ubuntu Precise) "No network support on Beagle XM" [High,Confirmed] https://launchpad.net/bugs/838200

 15:46:58 <rsalveti> sorry, I'm in two meetings

 15:47:00 <ogra_> [topic] AOB

 15:47:03 <rsalveti> GrueMaster: thanks

 15:47:04 <ogra_> anyone ?

 15:47:13 <GrueMaster> :)

 15:47:14 * ogra_ has nothing

 15:47:21 <infinity> I'm dry.

 15:47:41 <ogra_> seems nobody else either

 15:47:45 <ogra_> going once

 15:47:52 <ogra_> twice...

 15:48:02 <ogra_> sold to the guy in underwear ....

 15:48:05 <ogra_> #ednmeeting

 15:48:09 <ogra_> #endmeeting



Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/AlanBell/mootbot)