[00:49] <kees> hallyn, jdstrand: libseccomp 1.0.1 now in debian unstable and raring
=== fisted_ is now known as fisted
[03:48] <slangasek> smoser: bug #978127> OOI, what's the reason to not handle this by having the ephemeral image run ntp?
[03:48] <ubottu> Launchpad bug 978127 in cloud-init (Ubuntu Precise) "incorrect time on node causes failed oauth" [High,Triaged] https://launchpad.net/bugs/978127
=== Ursinha is now known as Ursinha-afk
[05:57] <infinity> tjaalton: BTW, I'm still getting i915 GPU hangs.  But they're no longer triggering apport popups.  I'm not sure this is progress. :/
[05:59] <lifeless> infinity: you're getting less apport errors, so the strategy is winning
[06:01] <infinity> lifeless: *smirk*
[06:01] <lifeless> :)
[06:01] <lifeless> Cheap shot, I know.
[06:03] <infinity> tjaalton: Readily reprodible (several times an hour) by watching The Walking Dead in mplayer.  I suspect it's the content that matters, turning my GPU into a zombie.
[06:03] <lifeless> nice
=== Ursinha-afk is now known as Ursinha
[06:03] <infinity> tjaalton: (It's also possible that my debugging skills and ability to tie cause to effect diminish slightly on weekends)
[06:07] <infinity> tjaalton: In all seriousness, though, there could be a bit of cause and effect here, as I can go hours working in terminals/firefox/etc and usually not tickle the bug, but it's never more than 15 or 20 minutes of x264 playback in mplayer before my GPU has a sad.
[06:30] <tjaalton> infinity: ok that's good to know, since I've been unable to make it hang myself (with normal "office" usage)
=== lifeless_ is now known as lifeless
[06:36] <tjaalton> infinity: also, are you able to reproduce bug 1086123
[06:36] <ubottu> Launchpad bug 1086123 in linux (Ubuntu) "3x power consumption compared to quantal" [High,Incomplete] https://launchpad.net/bugs/1086123
[06:36] <tjaalton> this after suspend/resume
[06:36] <tjaalton> haven't bisected it yet
[06:39] <infinity> tjaalton: Hrm.  I have no /sys/kernel/debug/dri ... Do I need to do something magical to get that?
[06:39] <tjaalton> guess you should see it as root?
[06:39] <infinity> tjaalton: (But I haven't noticed power consumption going up, I've been on AC since I got back from UDS)
[06:39] <tjaalton> heh, ok
[06:40] <infinity> (base)root@cthulhu:~# ls /sys/kernel/debug/
[06:40] <infinity> (base)root@cthulhu:~#
[06:40] <infinity> No such luck.
[06:40] <infinity> Did just get another:
[06:40] <infinity> [600472.021256] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung
[06:40] <tjaalton> debugfs is mounted here, dunno how to turn it on
[06:40] <infinity> tjaalton: Since apport's now failing to pick these up, how do I get more debug info than that for you?
[06:41] <tjaalton> infinity: "enable debugfs" would be the first step, but it's just working for me..
[06:42] <tjaalton> so I'm not sure how to enable it
[06:42] <tjaalton> i915_error_state would be there
[06:42] <infinity> Oh, hrm.  Maybe something's gone sideways here.
[06:43] <tjaalton> there were patches to make it root-only, but should still be available
[06:43] <tjaalton> this happened for precise I think
[06:43] <infinity> Yeah, there's an upstart job that makes sure it's read-only, but I can't find anything that's meant to actually mount it.
[06:43] <infinity> Perhaps mountall does it.
[06:43] <infinity> But clearly failed to.
[06:44] <infinity> Anyhow, manually mounted for now.
[06:44] <infinity> CAGF: 650MHz
[06:45] <infinity> Mine doesn't seem to be spinning at 1300.
[06:45] <infinity> I haven't rebooted into the -5- kernel yet, though.
[06:45] <infinity> So, that might break it.
[06:45] <infinity> Which would help your bisect.
[06:45] <tjaalton> another thing would be to give 'drm.debug=0x0e' to the kernel and then reproduce the hang
[06:46] <infinity> I'll have to give that a whirl when I'm not in the middle of being lazy at midnight. :P
[06:46] <tjaalton> infinity: have you suspended it?
[06:46] <tjaalton> sure
[06:46] <infinity> It's been suspended dozens of times since the last cold boot, I'm sure.
[06:46] <tjaalton> ok
[06:47] <infinity> Also, your bug report says "only the discrete GPU is in use", did you mean the inverse?
[06:47] <infinity> Or is your i915 spinning out of control despite you using the NVIDIA one?
[06:48] <tjaalton> I'm not using hybrid
[06:48] <tjaalton> oh
[06:48] <tjaalton> duh
[06:48] <tjaalton> fixed
[06:49] <tjaalton> this is slightly better with the current kernel
[06:49] <infinity> Oh, you're using -4.12- in that bug report too.  Same as me.
[06:49] <infinity> Weeeeeird.
[06:50] <infinity> Maybe we need to get our 420Ss to compare notes.
[06:50] <infinity> Or go to some sort of reform boot camp together.
[06:51] <tjaalton> hehe
[06:51] <tjaalton> with -5 it's slightly better
[06:52] <tjaalton> takes 22W, and 13W after fixing the stuck gpu
[06:52] <tjaalton> but still 5-6W extra
[06:52] <infinity> And every watt counts on this thing, with the teeny-tiny battery...
[06:53] <infinity> Anyhow, debugfs manually mounted, back to my TV.  I look forward to more apport popups again. :/
[06:53] <infinity> Maybe I'll reboot later.
[06:56] <tjaalton> cool, thanks
[06:57] <tjaalton> (i've got an extra battery mounted on the ultrabay ;)
[06:57] <tjaalton> gives 2-3h extra
[07:00] <infinity> tjaalton: Yeah, I've been considering that.  I'm miffed that their online custom ordering thing forces you to put a CD/DVD drive in the ultrabay, or I'd just order it with a battery straight up.
[07:00] <infinity> (Who needs optical media anymore?)
[07:04] <tjaalton> indeed..
[07:04] <vibhav> People still use it here :(
[07:09] <infinity> Alright, screw these hangs.  Rebooting.
[07:09] <infinity> tjaalton: drm.debug=0x0e you say?
[07:09] <infinity> tjaalton: What will that do?  Just more vomit in the ringbuffer?
[07:10] <tjaalton> infinity: more vomit in kern.log
[07:11] <infinity> tjaalton: That's what I said. :P
[07:12] <infinity> Haha.  Speaking of ringbuffer vomit, since I remounted debugfs, apport-gpu-erro broke.  Nice.
[07:13] <tjaalton> then when you get the hang, collect /sys/kernel/debug/dri/0/i915_error_state and attach to a bug (ubuntu-bug xorg, dmesg will be attached by it)
[07:14] <infinity> http://paste.ubuntu.com/1420467/
[07:14] <infinity> cat: /sys/kernel/debug/dri/0/i915_error_state: Cannot allocate memory
[07:14] <tjaalton> huh
[07:15] <infinity> My laptop is so confused right now.
[07:15] <infinity> We'll see how it fares after a reboot.
[07:47] <infinity> (base)adconrad@cthulhu:~$ ps ax | grep apport | wc -l
[07:47] <infinity> 22
[07:47] <infinity> Turns out this can bring a machine to its knees...
=== sraue_ is now known as sraue
=== Ursinha is now known as Ursinha-afk
=== tkamppeter_ is now known as tkamppeter
=== Ursinha-afk is now known as Ursinha
=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
=== cpg is now known as cpg|away
=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
[13:15] <jtaylor> infinity: you can sync cython from experimental
[13:15] <jtaylor> the ubuntu diff is not needed anymore
[13:15] <jtaylor> but did not test build it yet
[13:47] <voldyman> guys i want to modify/write-from-scratch the lock screen. is there any documentation/reference?
[13:49] <voldyman> anyone ? ^^^
=== voldyman_ is now known as voldyman
=== doko_ is now known as doko
[14:10] <jtaylor> where can I see why a package does not migrate from proposed?
=== Ursinha is now known as Ursinha-afk
=== voldyman__ is now known as voldyman
[14:19] <cjwatson> jtaylor: http://people.canonical.com/~ubuntu-archive/proposed-migration/
=== yofel_ is now known as yofel
=== fisted_ is now known as fisted
=== fisted_ is now known as fisted
[20:57] <blami> is there some description of unity? I don't understand if it's just couple of things built around gnome 2 compiz thing or gnome 3 shell ...
[21:02] <ScottK> blami: #ubuntu-desktop might be a better place to ask.
[21:03] <blami> ScottK: aha, thanks
[21:04] <Laney> More likely #ubuntu-unity ;)
[21:12] <ScottK> Even better.
=== cpg|away is now known as cpg
[21:40] <jtaylor> bdrung: I attached a patch to 1088051 which fixes the fftw uninstallablilty
[21:41] <jtaylor> seems to work fine in debian
=== mnepton is now known as mneptok