[00:01] <JanC> if I could only find where to look  :P
[00:01] <RAOF> You should just be able to start multiple X servers, IIUC.
[00:01] <JanC> each on its own monitor?
[00:02] <RAOF> Hm, probably.
[00:02] <RAOF> You'd want to read the fine Debian XRandR documentation to do that: http://wiki.debian.org/XStrikeForce/HowToRandR12
[00:16] <JanC> I don't see anything useful in there right now, unless not using a 2nd display in one X server makes it magically available in another one...  ;)...
[00:24] <RAOF> That's what I'd try.
[00:24] <Sarvatt> its not really documented anywhere yet, the X on wayland implementation is http://cgit.freedesktop.org/~krh/xserver/log/?h=hosted and http://cgit.freedesktop.org/~krh/xf86-video-intel/commit/?h=hosted&id=e0acdbd6f5f7ef8d9fc5b61fbd53041f4d74da06 but he said he's separating out the hosted stuff to allow X on X too
[00:24] <Sarvatt> hah! I just got a 24000 euro quote for a week stay at the hotel in budapest where the next UDS is
[00:25] <Sarvatt> $33,321, yeesh
[00:32] <RAOF> ‽
[00:34] <JanC> huh, and I thought the UDS conference hotel in Belgium was expensive...  ;)
[00:35] <JanC> Sarvatt: that's for 1 person?
[00:36] <Sarvatt> thats for the presidential suite for 2 people, it hid the cheaper options :)
[00:36] <Sarvatt> 249 euros a night, hopefully we get a good group rate so I can take the wife
[00:37] <JanC> I mean, for that sort of money they should at least provide you with 10 personal slaves or something  ;)
[00:39] <JanC> I remember when I went to Lisbon in 2001 or 2002, I had a room with a double bed, bathroom & sat TV for about 12-13 euro...
[00:39] <bjsnider> and they gave you 10 personal slaves?
[00:40] <JanC> no, not for 12-13 euro, but when comparing that to 24000 euro  ;)
[00:40] <JanC> it's nly 2000 × more  :P
[02:38] <ScottK> stgraber: re ltsp backports - Any reason we couldn't just put those in Ubuntu Backports once they are tested?
[02:40] <ScottK> RAOF: re the problem we briefly discussed the other day with a monitor coming back on saying "out of range" - I disabled powering off the monitor as you suggested.  Last night my wife came back to it and the monitor (LCD actually) was powered on and said "out of range".  Further suggestions?
[02:41] <RAOF> ScottK: Oh, wow.  That's weird.
[02:42] <ScottK> This is Lucid Kubuntu, btw.  It was stable on Karmic.
[02:42] <ScottK> (same hardware)
[02:42] <RAOF> So, my next thought would be that a screensaver is crashing X, corrupting the card state, and causing the respawning X to go mad.
[02:42] <ScottK> In which case I should find something about a crash in the logs.
[02:43] <RAOF> Yes.
[02:43]  * ScottK will have a look about for that then.
[03:44] <ScottK> RAOF: Would an xorg.conf that was left over from Jaunty that still called for Option "AccelMethod" "EXA" and Option "MigrationHeuristic" "greedy" be a potential source of "bad things"?
[03:44] <ScottK> No evidence of crashes, but I noticed that (along with setting dontzap) in the Xorg logs.
[03:44] <RAOF> Probably not.  EXA certainly wouldn't.
[03:45] <ScottK> That would just be a no-op at this point?
[03:45] <RAOF> Yes.
[03:45] <ScottK> I was thinking a reasonable next step would be to move the xorg.conf aside and see what, if anything, gets automatically generated.
[03:46] <RAOF> Nothing will get automatically generated, at least on the filesystem.
[03:46] <RAOF> Hm.
[03:46] <ScottK> OK, move it aside and see if it's absence helps.
[03:46] <ScottK> it's/its
[03:47] <ScottK> http://paste.ubuntu.com/529083/ is the non-comment parts of what's there.
[03:47] <RAOF> So, the out of range error suggests that something has triggered modesetting, and this modesetting has failed.
[03:47] <ScottK> OK.
[03:48] <RAOF> Oh, intel!   I don't think that AccelMethod option is going to take effect.
[03:48] <ScottK> Other than power state changes what might trigger that when it's not in use?
[03:48] <RAOF> Monitor hotplug, if it mis-detects a hotplug event.
[03:49] <ScottK> Hmmmm.  That would show up in dmesg?
[03:49] <RAOF> I can't really think of anything else.  Unless a screensaver is requesting a resolution change, but that would seem silly.
[03:49] <ScottK> Last time it happened I didn't personally verify the power was still on.  Maybe she was mistaken.  I'll keep an eye on it.
[03:50] <ScottK> In KDE there's more than one way to control monitor power state (of course) and I may have missed something.
[03:52] <ScottK> RAOF: Thanks for the feedback.
[03:52] <RAOF> Modesetting events won't turn up in dmesg unless you've got drm debug set; drm.debug=0x0e on the kernel commandline (or echoing to /sys/kernel/debug/dri/.../ at run time, I thikn) will set that.
[03:55] <stgraber> ScottK: yes, we could, though currently what we backported to lucid isn't even in natty yet (it's a bzr snapshot from upstream). I'd prefer to wait for something stable to hit natty first (with an actual upstream version number), then backport that.
[04:18] <RAOF> Well, that's pretty good.  30MiB on-disc saving in mesa!
[04:29] <RAOF> Now the obvious caveat: does it still work? :)
[04:32] <RAOF> Hm, and there's more to be had if gallium is added.
=== ion_ is now known as ion
=== kees___ is now known as kees
=== Bernardo is now known as Bernardo|Away
=== Bernardo|Away is now known as Bernardo
=== yofel_ is now known as yofel
=== seb128_ is now known as seb128
=== virtuald_ is now known as virtuald
[15:17] <ripps> does anybody know how make only 1 monitor my desktop, and the second only populated with widgets?
[15:37] <ripps> How about starting X only on my DVI monitor and leaving my VGA at a VT?
[15:38] <jcristau> can't do that.
[15:38] <ripps> why?
[15:39] <ripps> I just want a fullscreen terminal on my second monitor
=== Ian__ is now known as Ian_Corne
[18:32] <bryceh> ripps, ooh that'd be neat
[18:33] <bryceh> ripps, sounds like a feature that'd have to be implemented in the kernel
=== Bernardo is now known as Bernardo|Away
=== Bernardo|Away is now known as Bernardo
=== Bernardo is now known as Bernardo|Away
=== Bernardo|Away is now known as Bernardo