[13:37] <cr3> hi folks, the mouse is excruciatingly slow after starting ubuntu but seems to behave a few minutes later. does this sound like a problem with X?
[14:44] <mona-x> Sarvatt: i have a reproducible way of crashing X (xserver-xorg 1:7.6+7ubuntu7.1 from xorg-edgers). I need yet to narrow it down (it happens when I start firefox which has a lot of open tabs from last session, but it seems like restoring/reloading one specific tab triggers it). what's the best way to report this? I assume I cannot use apport/reportbug since this is not an official package?
[14:56] <mona-x> oops, actually this is an official package
[15:30] <mona-x> well, i got to go, sorry...
[17:31] <bryceh> cr3, there's been some changes in the input later the last couple months but I'd probably suspect the kernel first for something like that.  If you remember when you first started noticing it, boot a kernel from prior to that and see if it goes away.
[18:14] <cr3> bryceh: I've only been running Precise for a couple weeks, is that prior to the changes?
[18:16] <Sarvatt> sounds like the external outputs are being polled like a madman at startup like that bug from oneiric
[18:16] <bryceh> cr3, no there haven't really been many serious changes to the X stack the last couple weeks
[18:17] <bryceh> Sarvatt, could be
[18:17] <cr3> Sarvatt: is there a workaround I need to apply or should I expect this to be fixed shortly?
[18:17] <cr3> the problem is not so crippling, just not something I'd like to see endusers confronted with
[18:17] <Sarvatt> can you file a bug first? we can't tell without one :)
[18:18] <cr3> Sarvatt: ubuntu-bug kernel?
[18:18] <Sarvatt> ubuntu-bug xorg
[18:18] <cr3> Sarvatt: will do
[18:18] <bryceh> cr3, it's not likely to be a widespread bug
[18:18] <bryceh> (else we'd have gotten more bug reports lately)
[18:19] <bryceh> Sarvatt, wasn't that a displayport-related bug?  My memory's getting a bit vague on that
[18:20] <cr3> bryceh: this is a fresh install on a ThinkPad X200, so no display port in this case
[18:20] <bryceh> Sarvatt, will those probe pollings show up in Xorg.0.log or does he need to flip on drm.debug?
[18:20] <Sarvatt> phantom hdmi on some lenovo laptop i think, it was worked around in g-s-d in oneiric though and the bad interactivity at startup should be gone
[18:21] <Sarvatt> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/855124
[18:22] <Sarvatt> it was g-s-d probing it like crazy at startup and every probe killed desktop interactivity i believe
[18:22]  * cr3 reported bug #957298
[18:23] <bryceh> cr3, had you had an older ubuntu on the system previously?
[18:24] <bryceh> hah you do have displayport
[18:24] <bryceh> mm, gpu lockups
[18:26] <bryceh> actually, not a lockup.  EQ overflowing is probably a symptom of the lagging
[18:26] <cr3> bryceh: the laptop was running lucid before, I was only running oneiric in a vm, so I couldn't tell you whether this is a regression (even though I checked the regression checkbox)
[18:28] <bryceh> Sarvatt, not spotting evidence of excessive output polling in the Xorg.0.log
[18:29] <bryceh> and xrandr.txt not showing any phantom outputs left connected
[18:29] <bryceh> cr3, but one thing you could try is to set up an xorg.conf with snippets like this:
[18:29] <bryceh> Section "Monitor"
[18:29] <bryceh>         Identifier "HDMI1"
[18:29] <bryceh>         Option "Ignore" "True"
[18:29] <bryceh> EndSection
[18:30] <bryceh> do that for each port in your 'xrandr' output that you're not using
[18:30] <bryceh> if Sarvatt's right, then that should make it stop.
[18:30] <cr3> bryceh: how can I create the base xorg.conf?
[18:31] <bryceh> See https://wiki.ubuntu.com/X/Config for an example
[18:31] <bryceh> cr3, you can omit the Driver line
[18:33] <cr3> bryceh: sudo Xorg -configure returns a fatal error, am I supposed to run that from the console before logging in?
[18:34] <cr3> bryceh: or did you mean I should just put those monitor sections under /usr/lib/X11/xorg.conf.d/?
[18:35] <bryceh> cr3, no, scroll down to the Quick xorg.conf section, cut and paste that and delete the Driver line
[18:36] <cr3> bryceh: ah, my bad, didn't realize it was that simple :)
[18:37] <cr3> brb
[18:37] <Sarvatt> actually you may want to go /org/gnome/settings-daemon/plugins/xrandr in dconf-editor and try unchecking it and restarting
[18:38] <Sarvatt> bryceh: he's on the kernel with the phantom output probing time fix :(
[18:41] <bryceh> Sarvatt, yeah didn't spot evidence of that type of bug from the logs
[18:41] <cr3> bryceh: works! I updated the bug with the workaround you provided
[18:41] <bryceh> huh
[18:41] <Sarvatt> oh fun
[18:41] <bryceh> cr3, alrighty, well we definitely know how to fix this type of bug
[18:42] <cr3> bryceh: sweet, as long as your definition of fix is not creating an xorg.conf file :)
[18:42] <bryceh> cr3, hey one more favor - can you take another look at the wiki.ubuntu.com/X/Config page and see if the layout is clearer now?
[18:42] <bryceh> cr3, nope, it needs a quirk to the kernel
[18:47] <cr3> bryceh: definately clearer, the sudo Xorg -configure command might detract people when it would fail
[18:48] <bryceh> yeah, that's totally borked.  Was playing with it yesterday.
[18:48] <bryceh> for some reason Xorg -configure generates an xorg.conf that assumes there's a monitor on every probable output and configures them all.
[18:48] <cr3> bryceh: heh, that's likely to happen :)
[18:49] <cr3> bryceh: feel free to ping me again when the quick in the kernel is fixed, I could remove my xorg.conf file and provide feedback
[18:50] <cr3> s/quick/quirk/
[18:51] <bryceh> cr3, I posted a few more questions to the bug, that need answered first
[18:51] <bryceh> cr3, namely, try individually enabling each output and see which one(s) must be disabled to make the issue go away
[18:51] <bryceh> also, test a docking station if you have it
[18:52] <cr3> bryceh: yeah, just answered, you'll have to wait until I finish some work to test rebooting a few times :)
[18:52] <bryceh> no prob
[19:39] <FernandoMiguel> evening
[21:55] <bdmurray> bryceh: could you look at http://ubuntuone.com/2BsjpXA9nDVB1ZGd8FK646?
[21:57] <bryceh> looking
[21:57] <bdmurray> the is some "corruption" in the audacious window and right above the terminal
[21:57] <bryceh> wow, 3D text console
[21:58] <bdmurray> I think its out of focuse ;-)
[21:59] <bryceh> made a nice raised text effect ;-)
[21:59] <bdmurray> right below "Ever Neve" you can see what I'm talking about
[21:59] <bryceh> bdmurray, so, looks like no content retrieved for that particular attachment?
[21:59] <bdmurray> I'm concerned about the visual corruption of my windows
[21:59] <bryceh> anything unusual about the bug in question?  private or deleted attachment or something?
[22:00] <bdmurray> this is the X channel right?
[22:00] <bryceh> oh, the white bar on the bottom?
[22:00] <bdmurray> ;-)
[22:00] <bdmurray> the green line with dots around it
[22:00] <bryceh> aha, helps to zoom out
[22:00] <bdmurray> oh heh
[22:02] <bryceh> bdmurray, so looks like a fairly standard graphics corruption issue
[22:02] <bryceh> could likely be a driver bug.  Else might be compiz or even the app.
[22:03] <bdmurray> it happens with lots of apps
[22:03] <bryceh> ok, then that rules out app issue
[22:03] <bryceh> I've seen plenty of corruption issues like this before, but I don't think I've seen one reported recently
[22:04] <bryceh> if it can be repro'd in gnome shell or classic gnome (no effects), then that would absolve compiz
[22:04] <bryceh> bdmurray, when did it start doing this?
[22:04] <bryceh> bdmurray, and does it always happen or is there some steps to make it reproduce?
[22:05] <bdmurray> I don't have good steps to recreate it, it may have started last week
[22:05] <bryceh> booting a kernel from prior to last week might be useful then
[22:07] <bryceh> there haven't been any major changes to mesa lately, just a couple probably unrelated things, however mesa is sometimes to blame for these kinds of problems.  Last major update was 8.0.1 on Feb 23rd
[22:11] <bdmurray> okay, thanks for looking
[22:14] <bryceh> bdmurray, which driver is this?  -intel?
[22:16] <bdmurray> bryceh: radeon
[22:24] <bryceh> interesting
[22:24] <bryceh> bdmurray, I'm not seeing it on my radeon, but I've not updated in a bit.  I should do so...