[07:25] <mlankhorst> morning
[07:26] <tjaalton> yo
[08:31] <tjaalton> whoa, nice classic weave on resume with -proposed stack
[08:35] <mlankhorst> have to write linux/Documentation today :/
[08:37] <tjaalton> so sorry
[08:38] <mlankhorst> Nah it's ok, mostly copy pasting I hope. And since that is the only thing barring dma-buf-mgr right now. :)
[09:16] <mlankhorst> tjaalton: hey rc4 is out
[09:19] <tjaalton> mlankhorst: ooh
[09:20] <tjaalton> i pushed -intel 2.20.3
[09:20] <tjaalton> uploaded mesa 8.0.3-0u0.2 to precise-proposed
[09:20] <mlankhorst> erm most recent commit is 2.20.2-1ubuntu1?
[09:21] <tjaalton> should be uptodate
[09:21] <tjaalton> just that email is lagging behind
[09:21] <mlankhorst> ah k
[09:57] <mlankhorst> tjaalton: ok pushed rc4, but didn't test yet if it builds or not :)
[10:03] <mlankhorst> seems to build..
[10:06] <mlankhorst> and seems to start too, ship it?
[10:19] <tjaalton> yup!
[11:02] <tjaalton> oh right, I have the powers :)
[11:10] <akheron> mlankhorst: now running xorg-edgers and it still crashes :(
[11:10] <akheron> there are actually two kinds of crashes, one when suspending and the other when waking from suspend
[11:11] <akheron> when it crashes when suspending, the machine doesn't suspend, the X crashes and some services (like networking) goes down
[11:11] <akheron> and it suspends when I try to reboot, then when I wake it up, it reboots :)
[11:33] <tjaalton> looks like we missed the window for trivial bugfixes, duh
[11:38] <tjaalton> mlankhorst: you should include the debian changelog when merging from there, even if it's UNRELEASED ;)
[12:48] <mlankhorst> tjaalton: yeah think it got absorbed again :/
[12:48] <tjaalton> mlankhorst: what do you use to resolve the conflicts?
[12:49] <mlankhorst> automatic
[12:49] <tjaalton> automatic what?
[12:49] <tjaalton> git-mergetool?
[13:01] <mlankhorst> tjaalton: I forgot I set it up once and it auto merges changelog 
[13:02] <mlankhorst> and now I never have to worry about it again :)
[13:03] <tjaalton> mlankhorst: so you run git merge, and it complains about conflicts.. what do you do next?-)
[13:03] <tjaalton> it suggests git-mergetool, I use meld
[13:04] <tjaalton> as the mergetool
[13:12] <mlankhorst> i use.some debian changeloh merge tool
[13:13] <mlankhorst> automated it
[13:13] <tjaalton> but if you have other conflicts, those happen too
[13:13] <tedg> So, I'm in a bad place with that giant X update.  In fallback now, but I'm curious if there's an update planned, or I should try to back the whole thing out :-/
[13:13] <tedg> Guessing it might be something to do with the nvidia driver.
[13:13] <tjaalton> tedg: right, there is no nvidia driver yet
[13:13] <tjaalton> tseliot: ^
[13:13] <tjaalton> well, it has a wrong version
[13:13] <tedg> ?  Hmm, it updated?
[13:14] <tjaalton> oh?
[13:14] <tjaalton> dunno then
[13:14] <tedg> It was blocked for a while based on removing nvidia-common, and then it unblocked.
[13:14] <tedg> So I assumed that meant there was a proper version.
[13:14] <mlankhorst> tjaalton: no i fix most automated or follow xsf wiki
[13:14] <tseliot> tedg: what driver version are you using?
[13:15] <tedg> tseliot, 304.32-0ubuntu3
[13:15] <tjaalton> tedg: i guess the latest upload provides the new abi, so it'll install
[13:15] <tjaalton> something else wrong with it then
[13:15] <tedg> tseliot, I guess it'd be more correct to say I'm not using it :-)
[13:15] <tseliot> :)
[13:15] <tseliot> tedg: what's the problem, exactly?
[13:16] <tseliot> tedg: can't you install it or can't you get it to work once it's installed?
[13:16] <tedg> tseliot, Lightdm starts, as soon as I get to a Compiz desktop it crashes pretty quickly.  Unity-2d will run until I think something tries to use GL.
[13:16] <tedg> tseliot, My nvidia bug log is filled with: [   505.639] (EE) FBDEV(0): FBIOPUTCMAP: Invalid argument
[13:17] <Sarvatt> I never could get it working but RAOF said debian/changelog merge=dpkg-mergechangelogs in .gitattributes is supposed to work for changelogs
[13:17] <tseliot> tedg: it looks like you're using fbdev instead of nvidia
[13:17] <tedg> tseliot, It infact has that 429244 times
[13:17] <tjaalton> Sarvatt: it'll still fail when there are actual conflicts
[13:18] <tedg> tseliot, In the ~/nvidia-bug-report.log.gz
[13:18] <tedg> ?
[13:18] <Sarvatt> ah aka every time then :P
[13:18] <tjaalton> right, I guess it should still merge the old entries fine
[13:18] <tseliot> tedg: can you upload that file somewhere, please?
[13:20]  * tedg is now wondering if it pastebinit wasn't the best tool for this
[13:20]  * ogra_ found that the nvidia tegra driver has massive probs with the new cairo ... (1.10 works, 1.12 doesnt) probably their x86 drivers have a similar issue 
[13:20] <ogra_> the symptoms are very similar
[13:21] <tedg> tseliot, http://paste.ubuntu.com/1136092/
[13:21] <ogra_> i can start unity-2d, but as soon as i open anything or click an item it crashes ... unity 3d crashes immediately after login
[13:22] <tedg> ogra_, Hmm, yeah, does sound similar.
[13:22] <ogra_> but its arm vs x86 ... so you cant be sure, i guess the tegra driver still slightly differs from their normal binary driver
[13:26] <tedg> The bugs apport found are bug 1034378 and bug 1034379
[13:26] <ubottu> Error: Launchpad bug 1034378 could not be found
[13:26] <ubottu> Error: Launchpad bug 1034379 could not be found
[13:26] <tedg> Not sure they're really useful.
[13:26] <tedg> No, really, those are the numbers.
[13:26] <tedg> :-)
[13:35] <tedg> ogra_, Where's the best place to get Cairo 1.10?  Precise?
[13:36] <ogra_> uh, no idea :) i rolled back my driver instead of cairo ... 
[13:37] <ogra_> you might be able to get an older version on launchpad 
[13:38] <tedg> Yeah, I'd need to rollback all of the X stack to get the old driver... which will suck.
[13:47] <tedg> Okay, so the older version of Cairo means that GTK apps work better, but I still can't run Compiz.
[13:47] <tedg> So I'm guessing there's a GLX error and the new version of Compiz uses GLX.
[13:48] <tedg> s/Compiz/Cairo/
[13:48] <ogra_> fun
[13:49] <ogra_> does unity-2d at least work stable now ?
[13:49] <tedg> Yeah, it seems to.
[13:49] <ogra_> great so cairo is definitely involved on x86 too
[13:49] <tedg> I mean, as much as Unity-2d works :-)  The slow software sliding is annoying :-)
[13:49] <ogra_> stop it from sliding then ;)
[13:55] <tedg> ogra_, You know I can't do that ;-)
[13:55] <ogra_> hahaha
[13:56] <tedg> Okay, scratch that, Firefox brings the whole thing down.
[13:57] <tedg> Though, I think I can live without the web better than I can live without e-mail :-)
[13:58] <tjaalton> tedg: btw you have some ancient xorg logfiles making that dump huge :)
[13:59] <tedg> tjaalton, Are you calling me old?  ;-)
[14:00] <tjaalton> tedg: nah, the machine :)
[14:00] <tjaalton> logs from last year, one of which is huge
[14:00] <tjaalton> or was it older
[14:00] <tjaalton> just an observation
[14:01]  * tedg deletes them
[14:03] <ogra_> tedg, i'm not sure FF is even linked against the system cairo libs
[14:03] <ogra_> it might ship its own copy
[14:04] <tedg> ogra_, I think the bug is the first person to grab a GLX context looses.  FF is probably just the first person in this case.
[14:04] <ogra_> (chriscoulson shoudl be able to tell you)
[14:04] <ogra_> why would FF grab any GLX contexts ?
[14:04] <jcristau> because webgl
[20:02] <tjaalton> oh great, buggy -synaptics again.. bug 956071
[20:02] <ubottu> Launchpad bug 956071 in xserver-xorg-input-synaptics (Ubuntu) "Xorg crashed with SIGSEGV in XIGetDeviceProperty()" [High,Confirmed] https://launchpad.net/bugs/956071
[20:03] <tjaalton> jcristau: this is 681796 on debian
[20:04] <tjaalton> apparently reverting to 1.5.99.902-0ubuntu5 stopped the crashes
[20:05] <jcristau> sigh
[20:05] <tjaalton> so 1.6.0 is good, 1.6.2 is bad
[20:06] <jcristau> can touchpads stop existing already?
[20:06] <tjaalton> totally..
[20:12] <jcristau> tjaalton: so https://bugs.freedesktop.org/show_bug.cgi?id=52496 should be moved to synaptics?
[20:12] <ubottu> Freedesktop bug 52496 in Server/Input/Core "segfault in XIChangeDeviceProperty upon suspend" [Normal,Needinfo]
[20:14] <tjaalton> jcristau: yeah, that's what the testing suggests
[20:14] <tjaalton> I don't recall seeing that bug though
[20:14] <tjaalton> myself that is
[20:14] <tjaalton> but I trust keng-yu :)
[20:15] <jcristau> i'm not seeing anything related between 1.6.0 and 1.6.2
[20:15] <jcristau> oh well.  it's synaptics, i shouldn't be surprised.
[20:16] <tjaalton> hmm
[20:16] <tjaalton> let me check the backported patches first..
[20:19] <tjaalton> hum, didn't know we had 1.6.2 in precise-updates
[20:20] <tjaalton> so it's something between 1.5.99.902 and 1.6.0 that broke it
[20:38] <tjaalton> meh, git history lost
[20:41] <tjaalton> ah, no. just confusing commit msgs
[21:11] <mlankhorst> tjaalton: the real problem is that evil stack in 1.11 :(
[21:11] <tjaalton> mlankhorst: why is debian having the same bug?
[21:11] <tjaalton> with 1.12
[21:12] <mlankhorst> oh different bug, too many input bugs I guess :S
[21:12] <tjaalton> well, I'll ask yaroslav to test reverting to 1.5.99.902 or so
[21:14] <tjaalton> that lp bug was actually filed weeks before .902 hit precise, so dunno how many incarnations of a similar trace we have..
=== johanbr_ is now known as johanbr
[23:56] <Sarvatt> almost done fixing up the xorg-edgers piglit packaging to work for QA to use, will have daily builds ready soon if anyone needs it