/srv/irclogs.ubuntu.com/2008/01/17/#ubuntu-mobile.txt

bspencerif I want to get the source for cheese, how do I do that?00:38
bspencerwhat is the /etc/apt/sources.list value?00:38
amitkbspencer: add a new line with deb-src instead of deb. everything else remains the same00:40
bspencerdeb-src http://ports.ubuntu.com/ubuntu-ports gutsy main restricted universe multiverse00:40
bspencer?00:40
amitkshould be good00:41
bspenceror maybe deb-src http://archive.ubuntu.com/ubuntu ...   I'll try that00:41
amitkTeTeT: hey Torsten00:41
TeTeTamitk: Hi Amit, staying up late?00:42
amitkTeTeT: 3rd night in a row now00:42
TeTeTamitk: ouch, deadline approaching? 00:42
amitkTeTeT: how are you doing?00:42
TeTeTamitk: doing fine, currently in Taiwan, visiting partners00:43
amitkTeTeT: yeah.... trying to get graphics drivers on UME to behave with our kernel00:43
TeTeTamitk: I've read about poulsbo on the mailing list00:43
amitkTeTeT: good for you! You aren't going to be in London next week, are you?00:43
TeTeTamitk: no, I'm not coming for the sprint. Would I be needed there?00:44
amitkTeTeT: not really, just wondering if there is a coincidence00:44
amitkTeTeT: moving back to Germany soon?00:45
TeTeTamitk: I need to re-think my move back, relationship troubles00:46
amitkTeTeT: sorry about that00:47
=== mawhalen_ is now known as mawhalen
dholbachgood morning07:11
=== mdz_ is now known as mdz
=== doko_ is now known as doko
alek_deskamitk, there is a meeting today, right?11:56
=== asac_ is now known as asac
amitkalek_desk: yes there is11:56
alek_deskamitk, thanks.11:56
=== \sh_away is now known as \sh
=== cjwatson_ is now known as cjwatson
=== cjwatson_ is now known as cjwatson
Mithrandirhm15:01
* Mithrandir ponders how exactly to change ubuntu-meta for just mobile. I guess we can use the -mobile branch and point our PPA package to that.15:01
Mithrandircjwatson: ^^; it seems like update.cfg points to p.u.c/~cjwatson/seeds; that's just a seed mirror, right?15:08
Mithrandirso I should be able to point it to LP?15:08
* Hobbsee thought that was a redirect15:08
cjwatsonMithrandir: update uses --bzr by default, so we use the [hardy/bzr] section for that not [hardy]15:11
cjwatsonso in fact we use bzr+ssh://bazaar.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/15:11
Mithrandirahkay.15:11
cjwatsonif you need to change it for just mobile, I think I'd recommend a different source package15:12
Mithrandirwhy?  It'll be in a PPA.15:12
Mithrandircjwatson: ^15:15
Mithrandir(sorry if you're busy, just thinking you might be paying less attention to this channel than others. :-)15:15
cjwatsonoh, PPA, sure15:16
cjwatsonI don't care then :)15:16
Mithrandirok, good.  I'd be insane to do this in the main archive, agreed. :-)15:17
cjwatsonand indeed, I only really follow this channel when highlighted15:19
loolMithrandir: Is NM (daemon) running though?15:24
Mithrandirlool: yes15:25
* Mithrandir blames dbus15:25
loolMithrandir: Does restarting dbus help?15:26
Mithrandirno, that breaks things further; hald doesn't want to restart and it goes downhill from there15:27
loolAh got the hal issue too15:28
Mithrandirrebooting fixed that, though15:29
Mithrandirand stevenk's fix broke my hostname. :-/15:29
loolNot easy to fix when you can't sudo  :-/15:30
Mithrandirboot with break=bottom works, though15:30
* Mithrandir is lazy and just reinstalls15:33
Mithrandir*whine*; now I don't get any applets any more.16:02
Mithrandirah, unwhine.  It was just old packages.16:06
Mithrandirnm-applet works now that I twiddled the config slightly.16:06
Mithrandirit seems like it doesn't think we're on the console16:06
asacMithrandir: hmm ... that would explain the issue amitk complained about the other day16:07
Mithrandiryup16:07
asacMithrandir: do you have /bin/check-foreground-console ?16:08
Mithrandiryes16:08
asacand does it succeed?16:09
Mithrandirnjet16:10
Mithrandirgood $time_of_day everybody.16:56
Mithrandirdavidm,lool,amitk,agoliveira: around for the meeting?16:56
agoliveiraMithrandir: Yep16:56
amitkyup16:56
alek_deskhello16:57
Mithrandirhiya alek16:59
alek_deskhoya:) Mithrandir17:00
Mithrandir#startmeeting17:00
MootBotMeeting started at 17:00. The chair is Mithrandir.17:00
MootBotCommands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]17:00
agoliveiraCoup d'État? :)17:01
Mithrandirrustyl: is Mauri around and planning to attend?17:01
Mithrandirnah, davidm asked me to drive today.17:01
Mithrandirand since I've only had tea to drink, I think I'm fit.17:01
davidmagoliveira, yep what he said17:01
Mithrandirwell, tea and water.17:01
agoliveiraYou guys are boring :)17:02
rustyli'm online, doing multiple things... so slow response17:02
ToddBrandthere17:02
Mithrandirhiya Don_Johnson17:02
Don_JohnsonI'm here too.17:02
Mithrandirand hiya ToddBrandt, Bob17:02
bspencerhi17:03
davidmOK for the first open action we have ChickenCutlass here so we can go17:03
Mithrandirsure17:03
Mithrandir[TOPIC] Michael Frey to test the ext3 image boot speed before weeks end.17:03
MootBotNew Topic:  Michael Frey to test the ext3 image boot speed before weeks end. 17:03
MithrandirChickenCutlass: how did that go?17:04
ChickenCutlassTried it -- no faster17:04
davidmInteresting, would have expected some speedup.17:04
ChickenCutlasssame performance as shashfs17:04
ChickenCutlasssquashfs17:04
ChickenCutlasstakes up lots more space however :)17:04
Mithrandirnaturally.17:04
mdzChickenCutlass: is the startup process still CPU limited according to bootchart?17:05
Mithrandirdo you have bootcharts to compare the two?17:05
ChickenCutlassno -- I do not yet have bootcharts -- I did the stopwatch approach17:05
agoliveiraChickenCutlass: That's odd. There should be any improvment exactly due the CPU limitation I saw in the charts.17:06
davidmChickenCutlass, can you run bootchart on the ext3 boot so we can see exactly what is happening?17:06
ChickenCutlassI will build some charts so we can see17:06
Mithrandiragoliveira: your IO load goes up, though.17:06
mdzagoliveira: there's no particular reason to believe that squashfs is what's on the CPU, that was just a guess17:06
agoliveiraMithrandir: Exactly, maybe it freed the CPU but saturated the I/O17:07
smagounagoliveira: one theory was that squashfs required CPU, so load would go down when switching ti ext3.17:07
loolagoliveira: You have the URL to your bootchart available easily?17:07
agoliveiralool: No, sorry. I sent an email with them and deleted I guess.17:07
Mithrandir[ACTION] ChickenCutlass to produce boot charts for squashfs vs ext317:07
MootBotACTION received:  ChickenCutlass to produce boot charts for squashfs vs ext3 17:07
Mithrandirany other input on that topic, or move on?17:08
davidmagoliveira, where did you send the email I'll find it17:08
mdzmove on17:08
Mithrandir[TOPIC] bspencer to checked to see if get tags tagging has occurred when making a release.17:08
MootBotNew Topic:  bspencer to checked to see if get tags tagging has occurred when making a release. 17:08
davidmmove on17:08
loolhttp://diemicrosoftdie.com/gutsy-20071210-1.png 40417:08
MootBotLINK received:  http://diemicrosoftdie.com/gutsy-20071210-1.png 404 17:08
mdzget tags tagging?17:08
Mithrandircould somebody please explain that sentence?  It seems to lack some words or a spell check or something. :-)17:08
agoliveiradavidm: Right after the tests. I'll check later.17:09
bspencerwe talked about it 17:09
loolThe action item was presumably to check whether tags are actually happening for new releases17:09
Mithrandirlool: presumably, yes.17:09
agoliveiraAh... sorry, I sent an email with the links. I'll check the backups and try to put it up again.17:09
Mithrandirbspencer: have you checked?  and if so, what was the result?17:09
bspencerand plan to do this 17:10
bspencersorry...distracted17:10
bspenceryes, we have spotty coverage to date17:10
bspencerbut will tag these things as discussed previously17:11
bspencerok, 100% focused here now17:12
MithrandirI'll give you an action item to report back on whether it has improved over the next week17:12
bspencerwe want to tag the releases of each project when they are released17:12
Mithrandirok?17:12
bspencerMithrandir: sounds find17:12
bspencere17:12
Mithrandir[ACTION] bspencer to continue checking that projects are tagged when they release and report back on whether this is the case.17:13
MootBotACTION received:  bspencer to continue checking that projects are tagged when they release and report back on whether this is the case. 17:13
Mithrandirhiya Mauri17:13
Mithrandir[TOPIC] bryce and amitk to provide verbose emails to bspencer about issues surronding Tungsten. (complete)17:13
MootBotNew Topic:  bryce and amitk to provide verbose emails to bspencer about issues surronding Tungsten. (complete) 17:13
Mithrandirthis has happened, I believe.17:13
amitkyes17:13
bspencerI received many emails on the subject17:13
mawhalenMithrandir: hi17:13
Mithrandirhowever, we are still stuck with a broken -psb driver on Hardy.17:13
Mithrandiramitk: can you give us a bit more information on your findings?17:14
amitkBasically, while the psb X driver tries to acquire a DRI lock, it freezes17:15
Mithrandirare there anybody we could set up a conference call with, at intel, who could help us debug this problem?17:15
agoliveiradavidm: http://diemicrosoftdie.com/charts.html17:16
bspencerrob__: robr_ubuntu robr2  -- you on the call?17:16
amitkthe problem is in the kernel PSB driver according to current findings17:16
mawhalenMithrandir: Let me check with Rob Rhoads to see how much work he did with our 2.6.24 kernel.  17:16
MithrandirI see the same problem with the moblin 2.6.24 kernel, fwiw17:16
mawhalenMithrandir: some folks in our group are in another meeting right now17:16
mjg59I've some concerns about the interaction between the -psb driver and ACPI, but I believe this is a separate issue17:16
mawhalenis Don on?17:16
mjg59I'm looking at improving the in-kernel ACPI video support to the point where -psb can just use that17:16
amitkI have tried to use the moblin kernel configs on our kernel to see if there is a configuration issue, but the issue still remains17:16
mawhalenDon_Johnson: ping17:17
Mithrandiramitk: not really surprising given that the moblin kernels are broken too, at least for me.17:17
mawhaleninuka_desk: ping17:17
mdzmawhalen: should we reschedule?17:17
Don_JohnsonI'm here, just catching up17:17
mawhalenMithrandir: Inuka on our team is now responsible for the gfx integration.  I'll start inviting him to this meeting.17:18
Mithrandirmawhalen: which time zone is he in?17:18
Mithrandir(Inuka is a he?)17:18
mawhalenDon_Johnson: We might want to setup a meeting with Waldo, Rob and Amit.  Not sure if Waldo is still working on gfx.  17:18
mawhalenMithrandir: Inuka is a he, he is here in Oregon.17:19
mdzmawhalen: that's helpful for the future, though we would like to get this particular issue resolved sooner.  who can we talk to who has more familiarity with this driver?17:19
amitkcurrently I am digging into interaction of the X driver and the kernel driver. Someone at Intel who is familiar with these drivers would speed up the debugging a lot.17:19
Don_Johnsonmawhalen: I'll check with Waldo to see if we can set some thing up.17:20
Mithrandirideally, today.17:20
pat_mcgowanDon_Johnson, any chance Tobin could help out17:20
Don_Johnsonpat_mcgowan: Tobin isn't familiar with the internal workings of the graphics drivers.17:21
alek_deskMithrandir, but Moblin kernel 2.6.24 has a workable psb-kmd module17:22
Mithrandiralek_desk: doesn't work for me with hardy.17:22
alek_deskhmm...17:22
Mithrandir(verified two days ago)17:22
loolCould be a race triggerred by libdrm / xorg but only from hardy?17:22
MithrandirI haven't pried deeply into it, but it seems to get to the same point as the ubuntu kernel.17:22
bspencermdz: mawhalen and I will have to follow up to get someone who can help quickly.17:23
Mithrandirif you guys have something that works for you, it might be interesting to see what happens if we put our latest kernel in there and try.17:23
smagounmaybe a dumb question, does the hardy libdrm include intel's patches?17:23
Mithrandirsmagoun: this is with the libdrm I got from bryce which supposedly contains all the necessary patches.17:23
MithrandirI have not verified that, as I am not familiar with the code.17:24
Mithrandirshould we take this after the meeting?  It seems we're not making more progress here.17:25
davidmEarlier in the week bryce said he had applied all of the intel changes17:25
davidmYes, lets take it off line17:25
mdzis bryce not here?17:25
Mithrandirbryce: are you around?17:25
bspencerfrom our side let's grab waldo and rob for the discussion.  Inuka can come and listen too.17:25
mawhalenMithrandir: right, we don't have the correct people here17:25
mawhalenI can synch with Rob as soon as he is off his meeting17:26
Mithrandirok, good17:26
Mithrandir[TOPIC] mawhalen Don_Johnson davidm to meet off line on schedule issues (complete)17:26
MootBotNew Topic:  mawhalen Don_Johnson davidm to meet off line on schedule issues (complete) 17:26
Mithrandirthis is complete and done and nothing needs to be reported here?17:26
Mithrandirif so, the agenda is empty.17:27
Don_JohnsonYes, I believe it is done, and nothing needs to be reported here.17:27
Mithrandiranybody have anything more that needs discussion?17:27
Mithrandirgoing once, going twice.17:28
davidmDon_Johnson, your later call specific to the customer issues, correct?17:28
Don_JohnsonYes, we will be talking customer specific issues17:29
davidmOK, thanks17:29
Mithrandirok, thanks everybody.  Adjourned.17:29
Mithrandir#endmeeting17:29
MootBotMeeting finished at 17:29.17:29
Mithrandirthat was nice and short.17:29
agoliveiraBest... meeting... ever... :)17:29
amitkmawhalen: when can we know about who in Intel to talk to about the graphics drivers?17:29
Mithrandirmawhalen: when is rob's meeting done?17:29
mawhalenNot sure when Rob is done, I'll have him get ahold of amitk as soon as I find out.17:30
Mithrandirgreat.17:30
amitkthanks mawhalen17:30
GrueMasterSo, what's the key to getting a Hardy build in MIC?17:32
MithrandirGrueMaster: changing the default sources.list entries is at least one way17:34
GrueMasterOk.  Little background.  I have a cron that runs daily to pull MIC from git, then generate a daily Moblin-gutsy build.  It has the fsets for ubuntu-hardy as well.  What do I need to change (and why isn't it in MIC).17:36
bspencer> GrueMaster said:  d why isn't it in MIC17:38
bspencergood question.  If we're missing something useful send the maintainer a note or a patch17:38
MithrandirI believe the plan is to provide multiple platforms which default to different distroreleases17:39
loolATM, Ubuntu's MIC pulls only from Ubuntu while upstream's MIC pulls from Ubuntu + moblin.org's APT repos17:39
GrueMasteryes, and?17:39
GrueMasterso?17:39
GrueMasterActually, it ignores moblin.org if you select the ubuntu fsets only.17:40
loolThe fsets differ slightly in MIC upstream and in Ubuntu17:40
GrueMasterI understand.  Let's fix it so people using MIC can actually use MIC.17:41
Mithrandirrobr: just the guy we're looking for.  Can you talk to Amit about our current headaches on the kernel DRI bits?17:42
loolGrueMaster: I would be happy to work on taking all our fixes upstream17:43
loolGrueMaster: Would you like us to discuss the delta and address it now?17:43
cwong1Mithrandir:  I have an issue with flashplugin-nonfree and need to discuss with you. I was going to bring it up in the meeting but was chatting with Asac. 17:44
Mithrandircwong1: please shoot.17:45
amitkrobr: hi17:45
robramitk: hi17:46
cwong1Mithrandir: the current flashplugin-nonfree package has a bad checksum that caused the installation to failed.17:46
GrueMasterI just need to be able to get a working Hardy build so I can start addressing some of the build issues others are seeing in PSB17:46
Mithrandircwong1: hm, that's something we should fix17:47
loolGrueMaster: The easiest way would be to upgrade your hardy installation and use hardy's MIC17:47
amitkrobr: have you followed the thread on ubuntu-mobile ML? Subject: Backtrace17:47
cwong1Mithrandir: Asac has uploaded a fix a while back but have to rolled back. The reason is that the new flashplugin only work with Firefox/IE and will crash konque and opera.17:47
GrueMasterNo, that would break my other development systems17:48
robramitk: not yet, but i will try to catch up17:48
cwong1Mithrandir: Can asac upload the fix again? But this mean opera and konqer user will not be able to use the flashplugin.17:48
Mithrandircwong1: uh, ok.  It seems better to be able to install it on some, but not all systems, than on none?17:48
GrueMasterI don't have enough platforms to dedicate to everyone's individual changes that don't get sync'd mainstream.17:48
cwong1Mithrandir: I agree.  I will let asac know then.17:49
Mithrandircwong1: worst case, we can put it in the ppa17:49
cwong1Mithrandir: ok. thanks.17:49
amitkinuka: robr: Mithrandir: davidm: Can you all join #psb-hardy to continue this discussion?17:52
amitkbryce: GrueMaster: you too if interested ^^^17:56
asacMithrandir: will upload flashplugin to mobile ppa ... how does the versioning scheme look like?18:11
Mithrandirasac: append ~710um1 or ~804um1 18:12
Mithrandir(depending on gutsy or hardy)18:13
asacMithrandir: ok thanks18:22
asacMithrandir: when do you expect to go back to hardy?18:24
Mithrandirare you uploading this to the gutsy ppa or the hardy ppa or both?18:24
loolGrueMaster: I was told that perhaps what you wanted is to create a project / config in git MIC to pull from Ubuntu hardy + ppa; is that what you would like us to work on?18:25
asacjust <= gutsy is affected18:25
asacso just gutsy ppa18:25
Mithrandirasac: ok, thanks.  We're looking at moving to hardy, but gfx is currently busticated.18:26
GrueMasterWhat I need is either a saved project that someone else has generated, or the ability to generate a working (sans video driver) hardy image in MIC.  I also need to keep MIC pointing at moblin.org for moblin snapshot builds, plus gutsy snapshots for other people I support.18:27
GrueMasterMy current priority is to work on Beta 6 drivers for integration testing.18:27
asacMithrandir: is there any policy on what can be pushed to that ppa? or just "everything thats needed" ? e.g. could we upload latest xulrunner-1.9 and could we use system-cairo for reducing memory footprint (would require new libcairo)18:27
Mithrandirasac: all the stuff that's needed, yes.  Please make sure you don't break builds though18:28
Mithrandirmjg59: do you know how I can convince pam_foreground our mobile session actually is in the foreground?  My attempts at using openvt didn't really help.18:28
mjg59Mithrandir: Oh, I guess you never log in through pam...18:28
Mithrandirmjg59: I do su - ume -c ...18:28
mjg59Hm. Not off-hand. It's been a while since I touched that.18:28
Mithrandir'k18:29
=== mawhalen_ is now known as mawhalen
=== mawhalen_ is now known as mawhalen
=== mawhalen_ is now known as mawhalen
MithrandirStevenK: you broke ume-config-crown-beach when you called gconftool in the postinst without making sure gconf2 is installed first..20:36
StevenKOh, gah.20:36
StevenKMithrandir: Shall I'll fix the both of them (samsung-q1-ultra-config would be too) after the meeting?20:37
StevenKs/I'll/I/20:37
Mithrandirplease20:44
=== \sh is now known as \sh_away
=== inuka_ is now known as inuka

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!