20110127

December 16th, 2010, 15:00 UTC in #ubuntu-meeting.

Agenda

Action Items from last meeting

  • NCommander to talk to the release team on which team to track w.r.t. to bugs (co)

Standing Items

Meeting Outcome

[15:01] <NCommander> #startmeeting
[15:01] <MootBot> Meeting started at 09:01. The chair is NCommander.
[15:01] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[15:02]  * rsalveti waves
[15:02]  * NCommander waves
[15:02] <NCommander> who's here this morning?
[15:02] <janimo_> hello
[15:02]  * GrueMaster is not.
=== Ursinha-afk is now known as Ursinha
[15:03] <NCommander> GrueMaster: I know the feeling
[15:04] <NCommander> [link] https://wiki.ubuntu.com/MobileTeam/Meeting/2010/20110127
[15:04] <MootBot> LINK received:  https://wiki.ubuntu.com/MobileTeam/Meeting/2010/20110127
[15:04] <NCommander> [topic] Action Items from last meeting
[15:04] <MootBot> New Topic:  Action Items from last meeting
[15:04] <NCommander> [topic] NCommander to talk to the release team on which team to track w.r.t. to bugs (co)
[15:04] <MootBot> New Topic:  NCommander to talk to the release team on which team to track w.r.t. to bugs (co)
[15:04] <NCommander> c/o
[15:05] <NCommander> [topic] Standing Items
[15:05] <MootBot> New Topic:  Standing Items
[15:05] <NCommander> [link] http://people.canonical.com/~platform/workitems/natty/ubuntu-armel.html
[15:05] <MootBot> LINK received:  http://people.canonical.com/~platform/workitems/natty/ubuntu-armel.html
[15:05] <NCommander> [link] http://people.canonical.com/~platform/workitems/natty/ubuntu-armel-natty-alpha-2.html
[15:05] <MootBot> LINK received:  http://people.canonical.com/~platform/workitems/natty/ubuntu-armel-natty-alpha-2.html
[15:06] <NCommander> Burndown charts are looking better and are close to the trendline
[15:06]  * davidm is finally back hom
[15:07] <NCommander> morning davidm
[15:07]  * NCommander pokes everyone else
[15:07] <NCommander> no other comments on burndown charts?
[15:07] <ogra> Wi look okayish
[15:07] <ogra> i just closed my last one
[15:07] <NCommander> ogra: yay :-)
[15:08] <ogra> NCommander, you are actively working on yours i guess :)
[15:08] <NCommander> ogra: I hit a wall (more on that later)
[15:08] <NCommander> anyone else for comments?
[15:08] <ogra> did you talk to doko/linaro ?
[15:08] <ogra> rsalveti, are you likely to do all these package tests before A" ?
[15:09] <ogra> *A2
[15:09] <rsalveti> ogra: for qt, probably not, but I'm moving to the others now
[15:09] <rsalveti> lot easier
[15:09] <NCommander> ogra: not yet. I was waiting on confirmation on a maverick Qt 4.7.1 before I raise the 'its a compielr issue' alarm, but that built ICEd sometime in the middle of the night
[15:09] <ogra> and what about RAOFs items ?
[15:09] <rsalveti> for qt I need to rebuild all the packages that uses the gles backend
[15:09] <rsalveti> to see if it's going to break or not
[15:09] <rsalveti> and are all huge (kde ones)
[15:09] <rsalveti> so takes time
[15:09] <rsalveti> ogra: just moved to alpha 3
[15:09] <rsalveti> not important for now
[15:09] <ogra> great
[15:10] <ogra> GrueMaster, what about your checkbox tests ?
[15:10] <ogra> move or finish this week ?
[15:10] <GrueMaster> Will bump today.
[15:10] <ogra> wohoo
[15:10] <ogra> seems we look really good for A2 then
[15:11]  * NCommander reset his to INPROGRESS
[15:11] <NCommander> I keep forgetting we have that state
[15:11] <ogra> janimo_, if you need help with debian-cd NCommander and i have access to the builer machine
[15:11] <GrueMaster> ogra: Bumped to A3.
[15:11] <NCommander> fixed
[15:11] <ogra> though i must admit i'm not fully in the code since the crontab cleanup
[15:12] <ogra> validity checking for flavours changed
[15:12] <janimo_> ogra, thanks, I will probably ask for help in various forms
[15:12] <ogra> k
[15:12] <janimo_> but I need to figure out how to properly test locally
[15:12] <ogra> NCommander is probably deeper in the current code, but he will be busy with the QT issue
[15:13] <ogra> NCommander, move ?
[15:13] <NCommander> k
[15:13] <NCommander> [topic] http://qa.ubuntu.com/reports/team-assigned/canonical-arm-assigned-bug-tasks.html
[15:13] <MootBot> New Topic:  http://qa.ubuntu.com/reports/team-assigned/canonical-arm-assigned-bug-tasks.html
[15:14] <rsalveti> I have 2 that are already fixed released
[15:14]  * NCommander needs to fix that link
[15:14] <GrueMaster> I'll be working on a new list tomorrow F2F with bmurry.
[15:14] <ogra> NCommander, only if we have a new list
[15:14] <NCommander> likewise-open is an annoying one that reopened
[15:14] <NCommander> Wish it would stay closed for a cycle
[15:14] <ogra> (which is what GrueMaster refers to i think)
[15:14] <NCommander> will track that after Qt is fixed
[15:15] <ogra> NCommander, unassign :)
[15:15] <ogra> its desktop or server team
[15:15] <ogra> you shouldnt be assigned to it
[15:15] <rsalveti> ogra: any news about the qt package update?
[15:15] <rsalveti> to check the neon bug
[15:15] <NCommander> ogra: I've always been assigned to this one since its a porting bug
[15:15] <ogra> rsalveti, Riddell just told me it should be there any minute (two days ago)
[15:16] <Riddell> upstream are promising it soon but won't say any more
[15:16] <ogra> so i guess someone is working on it atm
[15:16] <rsalveti> hm, ok
[15:16] <ogra> NCommander, right, and you provided a patch, thats enough, package maintenance is supposed to be done by either desktop or server
[15:17] <NCommander> k
[15:17] <ogra> NCommander, find out which of the teams has the package and assign to the team
[15:17] <NCommander> will reassign
[15:17] <NCommander> server
[15:17] <NCommander> that I do know
[15:17] <ogra> i think its desktop
[15:17] <ogra> by say of mdz
[15:17] <ogra> (i remember a discussion when he lead the server team)
[15:17] <ogra> but ask pitti or robbiew
[15:18] <rsalveti> ogra: I think you still have the old alsa-utils bug
[15:18] <NCommander> k
[15:18] <rsalveti> but it's fix released
[15:18] <rsalveti> and it shouldn't
[15:18] <ogra> i do :(
[15:18] <ogra> its fix released for the kernel part
[15:18] <ogra> we should probably just open a new one for alsa-utils
[15:18] <rsalveti> ogra: hm, so we should open it or repen for the package
[15:19] <ogra> since that old bug got a lot of confusion in it
[15:19] <rsalveti> at least we're able to track it then
[15:19] <ogra> right
[15:19] <ogra> re-open then
[15:19] <rsalveti> janimo_: any news on the banshee bug?
[15:19]  * ogra saw some testing of the new upstream
[15:20] <ogra> that didnt look good
[15:20] <rsalveti> yeah, I remember something similar
[15:20] <NCommander> The underlying root cause is our mono stack is still foobar'ed. Someone needs to sitdown, and run the test suite failures
[15:21] <rsalveti> NCommander: you said you could do that in the past :P
[15:21] <rsalveti> as you seems to like fixing mono issues
[15:21] <NCommander> rsalveti: it keeps getting pre-empted by other things.
[15:21] <NCommander> also, like is the wrong word :-P
[15:21] <ogra> unity-2d vs QT is more important atm
[15:21] <rsalveti> sure
[15:21] <robbiew> ogra: FYI, while I believe desktop is the new owner of likewise-open...zul has resolved the issue and is uploading now
[15:21] <ogra> yeah, like is the wrong word
[15:21] <ogra> should have been LOVE :)
[15:22] <rsalveti> :-)
[15:22] <rsalveti> move?
[15:22] <ogra> NCommander, see, one item less on your plate ^^^
[15:22] <NCommander> ogra: don't make me hit you
[15:22] <janimo_> rsalveti, it happensa with 1.9.2 too
[15:22] <NCommander> robbiew: thanks for the confirmation
[15:22] <janimo_> I am almost sure it is a mono bug not banshee
[15:22] <rsalveti> weee :-)
[15:23] <janimo_> as the backtraces have  wildly different contents sometimes
[15:23] <janimo_> within CIL, SIGILLS, etc
[15:23] <rsalveti> must be fun to debug mono
[15:23] <ogra> thats just to entertain the debugger ;)
[15:23] <NCommander> janimo_: if you want to take a stab at it, and don't value your sanity, get me off meeting and we'll step through as far as I got
[15:23] <ogra> they add random messages ;)
[15:23] <janimo_> rsalveti, indeed, I do not yet know how to debug mono C#/C combination with gdb
[15:23] <janimo_> but I guess it's something I'll find out :)
[15:23] <ogra> i guess a talk with NCommander would help
[15:23] <NCommander> janimo_: I'm pretty sure the issue itself is in the C code with the trampoleans, etc.
[15:24] <ogra> he made some progress in the past
[15:24] <janimo_> NCommander, ok
[15:24] <rsalveti> yup
[15:24] <NCommander> [topic] Unity 2D Status
[15:24] <MootBot> New Topic:  Unity 2D Status
[15:24] <ogra> looks goood
[15:24] <NCommander> Broken on ARM :-/
[15:24] <ogra> MIRs filed
[15:24] <NCommander> ogra: thanks for handling that
[15:24] <NCommander> I'm running down the natty regresison
[15:24] <ogra> broken on arm, but not unity-2d's fault
[15:24] <NCommander> ogra: indeed, but its a bit of a wall banger
[15:24] <ogra> i hope it doesnt block the MIrs
[15:25] <NCommander> I've got a dialog open w/ linaro on it ATM, but probably going to need to run some more test builds of Qt
[15:25] <GrueMaster> Shouldn't.  It works on x86.
[15:25] <ogra> and i also hope the new QT might gain us something
[15:25] <ogra> but that might be a moot hope
[15:25] <NCommander> especially if it stoolchain
[15:25] <ogra> GrueMaster, a MIR requires that it builds and runs on all supported arches
[15:25] <rsalveti> ogra: I was thinking that the lack of proper translation would be worse for blocking mir than the seg fault issue
[15:26] <ogra> the translation has a milestoned upstream bug
[15:26] <rsalveti> as it's going to be used as the default ui for arm
[15:26] <ogra> should be fixed before natty
[15:26] <ogra> (botzh bugs are mentioned in the MIR)
[15:27] <rsalveti> yup, saw that
[15:27] <ogra> there goes your alter ego ;)
[15:27] <janimo_> NCommander, I can also look into QT debugging. I am just fresh out of a LibO debugging spree, so large C++ codebases failing in weird ways on arm no longer scare me as much as they used to
[15:27] <ogra> it would be good to have tracked it down before A2
[15:27] <NCommander> janimo_: you should look at the current dialog in #linaro
[15:27]  * janimo_ just found the cause of OO/Libo FTBFS
[15:27] <janimo_> NCommander, ok
[15:28] <ogra> since i would like to switch the images to unity-2d
[15:28] <NCommander> janimo_: isn't debugging OOo fun :-)
[15:28] <NCommander> ogra: I think realistically, we're looking at A3 for that
[15:28] <ogra> i guess libO makes it easier already
[15:28] <ogra> NCommander, well, wer should try at least
[15:28] <ogra> *we
[15:28] <ogra> we still have a week
[15:29] <NCommander> ogra: true, but I rather us cross the finish line with a well tested stack then push it into a2 at the very last possible moment
[15:30] <ogra> NCommander, if there are chances it works i'm fine getting images with untested stack
[15:30] <janimo_> NCommander, OTOH the sooner we push out thew more testers
[15:30] <ogra> right
[15:30] <ogra> thats what i mean
[15:30] <davidm> I agree with ogra and janimo_
[15:30] <ogra> but there needs to be at least a chance
[15:30] <NCommander> */2cents*
[15:30] <ogra> i wont switch if its clearly broken indeed
[15:31] <ogra> if its toolchain it would also help to have it identified before the new QT hits the archive
[15:31] <ogra> to prevent rebuilds
[15:31] <ogra> anyway, thats all from my side wrt unity-2d
[15:32]  * NCommander has nothing additional to add
[15:33] <NCommander> actually
[15:33] <NCommander> one thing
[15:33] <NCommander> I just got pointed at Qt's unit tests
[15:33] <NCommander> Will try and run with them after the meeting to confirm/deny a hosed Qt
[15:33] <ogra> ++
[15:33] <NCommander> (or more specifically, if I'm looking at a regression due to changed assembly)
[15:33] <ogra> the earlier we identify the better
[15:34] <ogra> move ?
[15:34] <rsalveti> move
[15:34] <NCommander> [topic] Kernel Status (cooloney)
[15:34] <MootBot> New Topic:  Kernel Status (cooloney)
[15:34] <rsalveti> NCommander: my name
[15:34] <ogra> and rsalveti :)
[15:34] <rsalveti> again
[15:34] <rsalveti> :-)
[15:34] <NCommander> [topic] Kernel Status (cooloney, rsalveti)
[15:34] <MootBot> New Topic:  Kernel Status (cooloney, rsalveti)
[15:34] <NCommander> ahem :-)
[15:35] <rsalveti> :-)
[15:35] <rsalveti> for omap 3 we just got an upload for a 38 based kernel
[15:35] <ogra> do we know where we stand wrt 2.6.38 on omap4 ?
[15:35] <ogra> and who finally does it
[15:35] <rsalveti> apw pointed the link to the already built deb, will give it a try on xm now
[15:35] <davidm> ogra, TI has just started working on it
[15:35] <rsalveti> ogra: for omap 4 that is still one open question
[15:35] <ogra> its nearly feb.
[15:35] <ogra> would be good to have something soon
[15:35] <davidm> and Linaro has just hired a gent to help
[15:36] <rsalveti> at least at the rally they said they were building the linaro landing team to work on that
[15:36] <ogra> ah, good
[15:36] <rsalveti> or at least to help on that
[15:36] <rsalveti> so work is probably being made
[15:36] <ogra> right
[15:36] <davidm> New Linaro gent is Andy Green
[15:36] <ogra> as long as we can get binaries before A3 i'm happy
[15:36] <davidm> Started work this week
[15:36] <ogra> i hope they dont build linaro packages but usable ubuntu packages
[15:36] <rsalveti> current ti status for the current kernel tree (35 based) is 4 releases ahead of us
[15:37] <rsalveti> mostly to make it work on ES2.2
[15:37] <ogra> (with security and SRU support)
[15:37] <rsalveti> but sebjan said it's still not stable, so not ready for merge
[15:37] <rsalveti> ogra: hm, don't know if we'll get it before A3
[15:37] <ogra> davidm, do you know anything about that ?
[15:37] <ogra> (SRU and security ... and also ubuntu-kernel compliance)
[15:37] <rsalveti> current upstream kind of work, but missing display drivers and a lot of other stuff useful for desktop
[15:38] <ogra> i'm a tad scared that they will only build linaro kernels
[15:38] <rsalveti> I'm not that sure
[15:38] <davidm> OMAP 4, I think it will be iffy to have it by A3 but I have some hope
[15:38] <davidm> We will get an Ubuntu kernel from TI/Linaro
[15:38] <ogra> good
[15:38] <rsalveti> yup
[15:39] <rsalveti> for Maverick, we got a new update for the main kernel
[15:39] <rsalveti> closing some quite old SRUs
[15:39] <ogra> finally
[15:39] <rsalveti> like bug 688765
[15:39] <rsalveti> and others
[15:39] <ubottu> Launchpad bug 688765 in linux (Ubuntu Maverick) "Can't init uart3 (no clocks available) at Beagleboard-xM" [High,Fix released] https://launchpad.net/bugs/688765
[15:39] <rsalveti> yeah, sru is taking quite some time
[15:40] <rsalveti> guess that's most of it
[15:40] <ogra> well, to keep tobin busy :)
[15:40] <rsalveti> brian is still working on the mem instability issue
[15:40] <GrueMaster> hey.
[15:40] <rsalveti> but no other news
[15:40] <ogra> move ?
[15:41] <rsalveti> move
[15:41] <NCommander> [topic] QA Status (GrueMaster)
[15:41] <MootBot> New Topic:  QA Status (GrueMaster)
[15:41] <GrueMaster> Testing daily unity-2d builds on maverick due to the QT natty issue.
[15:42] <GrueMaster> Testing other apps using the une-efl on natty.
[15:42] <GrueMaster> Other than that, waiting on replacement XM (shipping).
[15:43] <ogra> images work fine ?
[15:43] <ogra> i also think we can start ignoring une-efl bugs
[15:43] <GrueMaster> After the latest jasper-initramfs fix, so far so good.  Testing today's now to see if the une-efl session was fixed.
[15:44] <GrueMaster> I'm only using it until unity-2d is usable on natty.
[15:44] <ogra> right, indeed file bugs as you see them
[15:44] <ogra> i just mean fixing
[15:44] <ogra> isnt worth to put team energy into that
[15:44] <ogra> better focus on getting unity-2d fixed
[15:45] <GrueMaster> Not much I can do there.  I'm available to run tests though.  :P
[15:45] <rsalveti> yup
[15:45] <rsalveti> but that was the only main issue we had
[15:45] <rsalveti> and until having something with unity-2d, was good to fix
[15:45] <ogra> une-efl should be handled by community (speak up if you are community and intrested in taking it over :) )
[15:46]  * ogra hopes he made emmet happy with that ;)
[15:46] <ogra> move ?
[15:46] <GrueMaster> Until unity-2d is on the image, some support for une-efl needs to remain.
[15:46] <rsalveti> move
[15:46] <NCommander> [topic] ARM Porting/FTBFS status (NCommander)
[15:46] <MootBot> New Topic:  ARM Porting/FTBFS status (NCommander)
[15:46] <rsalveti> and janimo_
[15:46] <NCommander> <- *see previous Unity 2D discussion*
[15:46] <ogra> you do that alone now ?
[15:46] <NCommander> ugh
[15:47] <NCommander> need to update the dman termplate
[15:47] <NCommander> ahem
[15:47] <ogra> awesome
[15:47] <ogra> so janimo_ is free for other stuff
[15:47] <ogra> :P
[15:47] <janimo_> :)
[15:47] <rsalveti> :P
[15:47] <NCommander> [topic] ARM Porting/FTBFS status (NCommander, janimo)
[15:47] <MootBot> New Topic:  ARM Porting/FTBFS status (NCommander, janimo)
[15:47] <ogra> heh
[15:47] <NCommander> ogra: janimo_ is free, i want janimo :-)
[15:47]  * NCommander runs
[15:47] <ogra> i want both of them
[15:47] <ogra> four hands can work more than two ;)
[15:48] <ogra> any status ?
[15:48] <janimo_> some progress on haskell stuff
[15:48] <ogra> cool
[15:48] <janimo_> but lately takeon over by others
[15:48] <ogra> i saw a ton of syncs/uploads
[15:48] <janimo_> Laney and other haskelly people
[15:48] <ogra> yep
[15:48] <janimo_> LibO fixed wil hopefuly be in 3.3.1
[15:49] <Laney> should be fine
[15:49] <ogra> Laney, great work
[15:49] <janimo_> some NEW pending to unlock some other package to be asble to build darcs
=== bjf[afk] is now known as bjf
[15:49] <janimo_> and then the unavoidable failures because of swapping on the builders
[15:49] <ogra> sigh
[15:49] <janimo_> deferred until something with more than 512M is available
[15:50] <ogra> isnt that just timeouts ?
[15:50] <Laney> just ordering and stuff, you can follow on http://orangesquash.org.uk/~laney/haskell-installability/armel.png
[15:50] <ogra> i thought lamont has a whitelist for that
[15:50] <janimo_> ogra, yes, but I am not sure what the timeout is supposed to be if we don;t know how much it is swapping
[15:50] <janimo_> timeout was raised to 300 min from 150
[15:51] <janimo_> but a few packages are still taking longer
[15:51] <ogra> well, just double them with every build until you find a working solution
[15:51] <Laney> thanks for caring about universe stuff btw :-)
[15:51] <ogra> we need them buolt by release
[15:51] <ogra> *built
[15:51] <ogra> and its not clear when we get pandas
[15:51] <janimo_> ogra, I hope by releae we get at least one bettee build server
[15:51] <janimo_> even one [anda would help a lot
[15:51] <ogra> and even then the cluster isnt built yet
[15:52] <lamont> there's the other issue that when the machine goes through the swapstorm, launchpad sometimes decides that it's gone away and throws in the towel and goes to the next builder so it can timeout in several hours
[15:52] <ogra> bah
[15:52]  * lamont is working on those timeouts too
[15:52] <ogra> well, as soon as we get pandas they will go to davidm (i hope) and not to the DC
[15:52] <janimo_> ogra, you have an extra panda, you should hook it in the buildds ;)
[15:53] <ogra> i already sent my spare panda to lamont :P
[15:53] <janimo_> oh
[15:53] <ogra> but that was a pre-release and didnt work
[15:53] <lamont> ah.  grumichama
[15:53] <ogra> not sure that was ever worked around
[15:53] <lamont> if we haven't already, we can send that one back if you want it
[15:53] <ogra> nah
[15:54] <ogra> probably to davidm
[15:54] <ogra> i dont want it collecting dust here
[15:54] <GrueMaster> I think TI wanted the pre-2.0 8L boards back.
[15:55] <ogra> then david is the best bet to drop it off there
[15:55] <ogra> anyway, offtopic
[15:55] <rsalveti> and we're running out of time
[15:55] <ogra> move
[15:55] <lamont> janimo_: meanwhile, pester me later today and I'll bump the sbuild timeout before the next round of mass-giveback
[15:56] <janimo_> lamont, ok
[15:56] <ogra> NCommander, stop watching #linaro !
[15:56] <NCommander> [topic] ARM Image Status (ogra, NCommander)
[15:56] <MootBot> New Topic:  ARM Image Status (ogra, NCommander)
[15:56]  * janimo_ set up a timer task titled 'pester lamont'
[15:56] <ogra> they are there ;)
[15:56] <NCommander> ogra: that was a case of bad lad
[15:56] <NCommander> *lag
[15:57] <ogra> x-loader was moved in the scripts to the new packages
[15:57] <ogra> as soon as i have confirmation they actually ended up on the images we can remove the old packages
[15:57] <rsalveti> yup, doing that now
[15:58] <ogra> beyond that we had a mis-build due to archive skew two days ago
[15:58] <ogra> all other images up to today have built fine
[15:58] <ogra> move ... unless there are questions
[15:58] <ogra> NCommander, HURRY UP !!
[15:58] <ogra> :)
[15:58] <rsalveti> ..?
[15:58] <rsalveti> :-)
[15:59] <ogra> NCommander, 2 MINS
[15:59] <ogra> hrm
[15:59] <ogra> my caps key freaks out
[15:59] <rsalveti> GrueMaster: give him a hard ping
[15:59] <ogra> with a hammer
[15:59] <rsalveti> lol
[15:59] <NCommander> [topic] AOB
[15:59] <MootBot> New Topic:  AOB
[15:59] <ogra> nothing here
[16:00]  * NCommander has nothing
[16:00]  * NCommander needs to reboot badly
[16:00] <NCommander> closing meeting in
[16:00] <NCommander> 1
[16:00] <ogra> NCommander, close ?
[16:00] <NCommander> 2
[16:00] <ogra> ah
[16:00] <NCommander> 3
[16:00] <GrueMaster> Actually, try not to reboot badly.  Makes your system slow down.
[16:00] <NCommander> #endmeeting
[16:00] <MootBot> Meeting finished at 10:00.

Action Items

  • None

Weekly Reports

Minutes

ARM/Meeting/2010/20110127 (last edited 2011-07-28 17:59:07 by davidm)