[00:50] <wgrant> 2
[03:38] <Sarvatt> huh...
[03:38] <Sarvatt> The following packages have unmet dependencies:
[03:38] <Sarvatt>   libgl1-mesa-dri: Depends: libdrm-radeon (>= 2.4.12) but it is not installable
[03:38] <Sarvatt> E: Broken packages
[03:39]  * Sarvatt didnt touch the control files
[03:39] <bryce> man goes into car dealership
[03:39] <bryce>  "I bought a car from you, but now it doesn't work"
[03:39] <bryce>  "Oh, sorry to hear that.  What's wrong with it?"
[03:39] <bryce>  "Well, I try to drive it, but it doesn't go."
[03:40] <bryce>  "That's a pity, can you give us some more information?  Or just bring it in and we'll take a look at it."
[03:40] <bryce>  "But I can't bring it in, it doesn't go!"
[03:41] <bryce> ^-- how I feel things are going with bug #399070
[03:41] <Sarvatt> ahh crap, i bet it pulled the libdrm-radeon from the new libdrm-radeon1.symbols i made
[03:55] <Sarvatt> you summed it up pretty well there :)
[03:57] <Sarvatt> well that was a good time for people to test out the script since edgers was broken for a bit :D
[03:58] <Sarvatt> made a new libdrm-radeon1.symbols since i finally figured out how and didnt change the first line -- libdrm_radeon.so.1 libdrm-radeon #MINVER#
[04:04] <Sarvatt> xorg is going to be a close call, fedora is for sure using 7.5 in F12 but i think they're releasing a month after karmic
[04:06] <bryce> yeah they also are more directly involved upstream so have more influence over the actual timing of the release
[04:06] <bryce> so unfortunately often things get timed more conveniently for them than for us...
[04:07] <bryce> plus, they seem to be a bit more liberal in the level of upstream QA they accept
[04:09] <Sarvatt> would have to drastically diverge from debian i'd think at this point to go 7.5
[04:15] <Sarvatt> i cant keep up with it, need to rebuild large chunks of the xorg stack every week because of all of the changes
[04:16] <Sarvatt> oh, fedoras final release is november 2nd..
[04:22] <Sarvatt> hmm every single person running 1024x600 posting traces of usplash crashes has the same exact thing, starting to wonder if it really is just this resolution screwed up
[04:22] <Sarvatt> recompiled it with debug info and uncommented the printf's around where it segfaults http://launchpadlibrarian.net/29680345/usplash_bt_dbg.txt
[04:26] <bryce> whew long day... -intel's down to 155 bugs.  I'd keep going but I should probably stop working and go entertain the wife.  cya.  Tomorrow I switch my focus to -ati.  :-)
[04:35] <Sarvatt> see ya bryce! yeah I need to do the same, shes talking my ear off and i'm not paying attention :)
[08:10] <tjaalton> there's something seriously wrong about virtualbox, because there are a ton of crash reports flowing in
[11:16] <RAOF> Sarvatt: Oooh.  Your libdrm to the PPA upload broke nouveau.  I'll just upload a newer snapshot, that'll work.
[14:51] <Ng> I guess it'll be a while before XInput2 is widely deployed?
[14:52] <Ng> I was about to poke around some of the recpies, but I realised I don't have it in karmic ;)
[15:08] <tjaalton> right
[15:09] <tjaalton> xserver bugs down to 199 (from 220+)
[15:14] <Ng> :)
[15:14] <Ng> have there been any hug days for X bugs? every now and then I look over all the bugs relating to chipsets I own, but they're rarely things I've ever been able to reproduce or test :/
[15:16] <tjaalton> I think there's been one
[15:19] <tjaalton> but they don't fix that well, since most of the bugs have been touched already
[15:19] <tjaalton> uh, "fit"
[15:36] <tjaalton> sweet, I could reproduce the crash in vbox
[18:57] <bdmurray> bryce: do you have a bug like bug 406996 - black lines on screen after logging in until desktop fully loads?
[18:59] <bryce> bdmurray, nope
[18:59] <bryce> bdmurray, guessing its some bug with the mythtv login dialog?
[18:59] <bdmurray> bryce: I've seen it too on my laptop
[18:59] <bryce> bdmurray, it sort of looks like a shadow rendering error
[19:00] <bryce> what's your laptop got installed?
[19:00] <bryce> bdmurray, tell me more about your laptop
[19:01] <bdmurray> bryce: Ubuntu! its the new login dialog as soon as I click my name and it shrinks I see the black bars
[19:03] <bdmurray> bryce: it has intel 945GM/GMS
[19:04] <bryce> with or without compiz?
[19:04] <bdmurray> with compiz
[19:04] <bryce> does it repro if compiz is disabled?
[19:05] <bdmurray> is disabling compiz enough or do I need to restart gdm or something?
[19:05] <bryce> I think disabling compiz and logging out / in should be sufficient
[19:06] <bdmurray> okay then it still happens with compiz disabled
[19:46] <bryce> bdmurray, ok, run apport-collect 406996 and I'll just upstream it
[19:47] <bryce> maybe it's not an X bug, but if not I'm not sure what'd cause it.  seems coincidental both you and the other reporter have i945.
[19:56] <Sarvatt> bdmurray: funny, i cant reproduce that with the same exact machine
[19:57] <Sarvatt> that seems like a mythbuntu problem with the special hooks in gdm it has
[20:09] <Sarvatt> wow good results from the new pixman on armel https://bugs.edge.launchpad.net/ubuntu/+source/pixman/+bug/385553
[20:09] <Sarvatt> a large portion of the results are almost 2x faster
[20:33] <bryce> bdmurray, hmm folks were talking about enabling compiz on the gdm login screen at one point, I wonder if that's what's going on
[20:34] <bryce> you mentioned the login box "shrinks" - is that a "compizy" type shrinking?
[20:40] <Sarvatt> you can edit that in gconf-editor at /apps/gdm/simple-greeter
[20:41] <tjaalton> it's the new login box that does that. there's no compiz running until the session has loaded
[20:41] <Sarvatt> you can make it use compiz for the GDM session, that might even be the default behavior now
[20:41] <tjaalton> not on my laptop
[20:42] <tjaalton> although maybe I didn't reboot it after today's updates
[20:42] <Sarvatt> i dont use compiz so i have no idea if it picks that if its enabled on upgrade or something
[20:43] <bryce> tjaalton, are you also seeing the line droppings?
[20:43] <Sarvatt> it looks like the greeter box doesnt fully clear, thats the size of the box
[20:43] <tjaalton> bryce: haven't seen them
[20:43] <tjaalton> but then again I'm not sure if something changed very recently
[20:44] <bryce> well, I forwarded the issue upstream anyway - https://bugs.freedesktop.org/show_bug.cgi?id=23047
[20:44] <tjaalton> ah, those stripes
[20:44] <Sarvatt> it doesnt happen here on the same machine as his, thats why i'm guessing its a mythbuntu specific setting since mythbuntu has a bunch of custom things done to gdm to get around the greeter
[20:44] <tjaalton> I saw them with virtualbox :)
[20:44] <bryce> Sarvatt, well bdmurray says he sees it and he runs stock ubuntu
[20:44] <tjaalton> which is not using intel
[20:44] <tjaalton> I'd say it's gdm
[20:45] <Sarvatt> bdmurray's picture shows mythbuntu in the menu?
[20:45] <bryce> yeah I speculated similarly on bug #406996
[20:45] <bryce> Sarvatt, that's not bdmurray's picture
[20:45] <Sarvatt> oh
[20:45] <tjaalton> well, vesa or vboxvideo both don't support compositing
 bryce: Ubuntu! its the new login dialog as soon as I click my name and it shrinks I see the black bars
[20:46] <crevette> know problem, csw problem on gtk
[20:46] <crevette> uptream bug is laready opened
[20:46] <bryce> crevette, oh... got a bug id?
[20:46] <crevette> bryce, let me find it
[20:46] <bryce> thanks
[20:46] <Sarvatt> i thought https://bugs.edge.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/406996 was his bug, sorry
[20:46] <crevette> csw on gtk+ caused few problem
[20:46] <maxb> Can anyone tell me what happened to the i810 video driver after the 2.4.x series? 
[20:47] <tjaalton> maxb: it's been -intel since 2.0
[20:47] <crevette> bryce, http://bugzilla.gnome.org/show_bug.cgi?id=589369
[20:47] <bryce> crevette, thanks
[20:47] <tjaalton> i810 has been removed from the archive since hardy I think
[20:47] <Sarvatt> ahh so disabling the background plugin in gdm might fix it, that would have been the first thing i'd try :D
[20:47] <jcristau> xserver-xorg-video-i810 | 2:1.7.4-0ubuntu7 |         hardy | source, amd64, i386
[20:47] <jcristau> xserver-xorg-video-i810 | 2:2.4.1-1ubuntu10 | intrepid/universe | all
[20:48] <tjaalton> oh right
[20:48] <crevette> bryce, for csw problem you want to look at this query http://tiny.cc/Gwf3d
[20:48] <tjaalton> it was in hardy, and intrepid had the dummy
[20:48] <crevette> or do it yourself by putting csw in whiteboard field in bugzilla
[20:49] <jcristau> tjaalton: won't you need a dummy for upgrades to the next lts btw? or does update-manager take care of making sure -intel is installed?
[20:49] <crevette> csw = Client Side window for those not aware 
[20:49] <tjaalton> jcristau: I think it does. we'll probably see before too long :)
[20:50] <tjaalton> although, didn't someone from intel say that -i810 should be used for i810 chips?
[20:50] <tjaalton> which would mean to revive the driver \o/
[20:50] <jcristau> well
[20:50] <bryce> bdmurray, ok there we go then.  known gdm issue.
[20:50] <bdmurray> bryce: yes, I see that now
[20:50] <crevette> not gdm
[20:51] <jcristau> i don't think i810_driver.c has changed that much from the earlier driver
[20:51] <crevette> gtk problem
[20:51] <bdmurray> sorry about that then
[20:51] <Sarvatt> gconftool-2 -s '/apps/gdm/simple-greeter/settings-manager-plugins/background/active' --type bool false
[20:52] <Sarvatt> ?
[20:52] <tjaalton> jcristau: true, and I doubt there are many laptops with that hw working anyway
[20:52] <Sarvatt> they even dropped i810 from xf86AutoConfig.c in xserver master
[20:52] <tjaalton> them becoming older
[20:53] <tjaalton> about time
[20:53] <bryce> I've run into a couple i810 users this year, one of which was at UDS
[20:53] <tjaalton> people are getting worried because the logs have error entries about the driver missing
[20:53] <tjaalton> bummer :)
[20:54] <jcristau> so for an i810/i815 the current driver will work as well/bad as the old one probably
[20:54] <Sarvatt> i've got that background plugin in gdm disabled here because it makes the screen flash loading the second background plugin after you login
[20:54] <jcristau> except the old one wasn't updated for pciaccess and other abi changes, so.
[20:54] <tjaalton> ah, right
[20:54] <bryce> very recently I had a guy approach me about an i830 bug in a laptop he *just* bought.  I had to ask how much he paid!  ($75 on ebay it turned out)
[20:54] <tjaalton> so no problem there then
[20:55] <tjaalton> poor man
[20:55] <tjaalton> not literally :)
[20:55] <bryce> hehe
[20:55] <bryce> he said windows xp worked fine on it, so I wished him luck with that
[20:55] <tjaalton> haha
[20:56] <bryce> (but I did kindly make sure the bug got forwarded upstream after advising him it probably would not get attention upstream)
[20:57] <tjaalton> having this jerrylamos guy around might prove very useful
[20:57] <tjaalton> with all the old hardware
[20:57] <bryce> yep, I've been pleased with the bug reports he's been filing
[20:59]  * Sarvatt wonders why he doesnt just use vesa since he seems to think gtkperf compared against vesa = a real world performance benchmark
[21:01] <tjaalton> I think he's genuinely interested in seeing the OOTB-experience to be "good enough"
[21:02] <Sarvatt> OOTB experience playing a gtk port of progress quest maybe :D
[21:02] <tjaalton> well I don't know the specifics :)
[21:02] <tjaalton> superm1: can you disable the other nvidia device from the XPS 1340?
[21:03] <Sarvatt> #3 0x00871096 in usplash_clear (x1=332, y1=457, x2=692, y2=607, colour=0) at libusplash.c:228
[21:04] <Sarvatt> ahhhhh my y res is only 600
[21:04] <tjaalton> Sarvatt: what does /etc/usplas.conf say?
[21:04] <tjaalton> +h
[21:04] <Sarvatt> 1024x600
[21:05] <tjaalton> ok
[21:05] <Sarvatt> themes screwed up
[21:05] <Sarvatt> it doesnt use 1024x600 in non KMS so i didnt notice
[21:06] <superm1> tjaalton, no that's not possible
[21:08] <tjaalton> superm1: ok, that would've been handy to test if the vesa probing fails because there are two devices
[21:09] <superm1> tjaalton, that wouldn't surprise me.  i believe that you need to start from the primary device
[21:13] <Sarvatt> if i had one of those laptops, i would want to always use the dedicated video in linux since the alternative is never using the dedicated video and always using the IGP.. thats how the vaio bioses do it at least, but acpi checks for linux osi dont do anything since it returns vista so you have to force it on the kernel command line
[21:15] <Sarvatt> can you change the primary video device on the xps 1340 in bios?
[21:15] <Sarvatt> not disabling the other just making the 9200 primary
[21:16] <Sarvatt> things should work if you could do that, the 9200 works fine with -nv
[21:22] <superm1> Sarvatt, i dont believe you can force the primary with the way it's wired (at least from what I last heard from NV)
[21:22] <bryce> Ng, mind upping the priority on your fdo bug 22186 to high please, so it matches the priority in lp?
[21:22] <superm1> there isn't a setting in any of the shipping bios'es for it (development bioses used to have that feature on the early versions of the hardware)
[21:22] <Sarvatt> ohh darn :( 
[22:42] <Ng> bryce: done :)
[22:43] <bryce> Ng, thanks
[22:44] <Ng> bryce: I noticed a similar thing today, if I have the mouse inside vinagre's VNC widget the screensaver doesn't seem to kick in, but moving the mouse out activates it almost immediately
[22:45] <Ng> I need to check that I'm not mistake and work out a simple test case before I file it though
[22:48] <virtuald> have you found a reason why the edgers xorg or its ati driver turns off the displays at random? like some inactivity counter doesn't stop counting sometimes when it should
[22:51] <bryce> it's not edgers, it occurs with stock ubuntu too
[22:51] <bryce> I've been noticing that issue myself for quite some time
[22:51] <bryce> my guess is it's some quirk with power management or the screensaver, but I've never chased the issue down
[22:52] <bryce> seems to happen a fixed time after waking up the displays.  15 min?  I've not done an exact measurement
[22:52] <virtuald> i don't know much c but can i do anything to help anyway?
[22:52] <virtuald> hm i don't remember a pattern like that yet
[22:52] <bryce> virtuald, I think a first step would be to try to figure out a reliable way to reproduce the issue
[22:53] <virtuald> ok
[22:53] <virtuald> also should this: http://launchpadlibrarian.net/29669968/OopsText.txt be upstreamed?
[22:53] <virtuald> and to who?
[22:54] <bryce> my guess is that if you put the monitors into power saving mode, then woke them up and noted the time, and then when it blanks note the time at which that occurs
[22:54] <bryce> repeat that a few times to see if the time is the same
[22:55] <bryce> virtuald, that looks like a kernel oops, so should be filed against the kernel
[22:55] <virtuald> ok
[22:55] <bryce> if you think there's an X-ish aspect to it, you can tag it 'xorg-needs-kernel-fix' so it gets into the kernel/X queue
[22:56] <bryce> the 'radeon_ttm_*' stuff seems to indicate that's the case
[22:56] <virtuald> yeah
[22:56] <virtuald> i get it on every boot
[22:56] <virtuald> should ad that too
[22:57] <virtuald> is it ok if i just update the description and not add a new comment? does the right people get pinged with that tag alone?
[22:58] <bryce> yes
[22:58] <bryce> it's not a ping, but rather causes the bug to show up in a certain report
[22:58] <virtuald> ok
[23:00] <virtuald> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/406478