[00:00] <ubotu> New bug: #189772 in mesa-utils (main) "[hardy] glxinfo uses 100% CPU" [Undecided,New] https://launchpad.net/bugs/189772
[00:05] <ubotu> New bug: #150638 in linux-source-2.6.22 "framebuffer mode results in black screen starting with kernel 2.6.22-13-generic (dup-of: 129910)" [Undecided,New] https://launchpad.net/bugs/150638
[02:49] <ubotu> New bug: #183869 in xorg-server (main) "mouse is not working after upgrading to hardy" [High,Confirmed] https://launchpad.net/bugs/183869
[07:22] <tjaalton> bryce: whoa, you closed a couple of intel bugs :)
[07:23] <bryce> :-)
[07:24] <bryce> tjaalton: halfway to my goal of reducing -intel to <100 bugs
[07:25] <tjaalton> I guess there are a lot of bugs on drivers which should really be against mesa
[07:26] <tjaalton> DRI, 3D etc
[07:26] <bryce> could be
[07:26] <bryce> it's interesting how much easier it is working on bugs once they're cataloged properly
[07:27] <tjaalton> right
[07:27] <bryce> so like today since I'd finished with New, I just focused on bugs marked 'Incomplete'
[07:28] <bryce> next time I'll focus on the confirmed ones and get them upstreamed
[07:28] <tjaalton> or ask to try with hardy
[07:28] <bryce> I did that a lot today
[07:28] <tjaalton> yeah
[07:28] <bryce> it was interesting just how large a portion of the bugs were plain old out of date
[07:30] <bryce> sort of makes me think how unproductive a lot of the bug reporting efforts are - the bug report never gets seen upstream, yet the issue got fixed anyway, so the effort of reporting the bug merely generated work, with no benefit.
[07:30] <bryce> not really their fault though of course
[07:31] <tjaalton> I think our documentation should be changed to use 'lspci -nn | grep VGA', which should be enough. -vvnn creates too much noise
[07:32] <tjaalton> thats partly because there are too many bugs open
[07:32] <tjaalton> ..still
[07:32] <tjaalton> so if we only had those that we are certain about, it would be easier to identify new ones and forward upstream
[07:34] <bryce> yup
[07:36] <tjaalton> having a release biannually doesn't really help :P
[07:37] <bryce> in any case, I feel like we're making good progress with them
[07:38] <bryce> I'd like to get to a point where I can focus on creating fixes more than just managing bug loads
[07:57] <tjaalton> a reasonable goal indeed :)
[07:58] <tjaalton> yay, we are below 1600 bugs (1580)
[07:59] <bryce> nice; where is that totaled?
[07:59] <tjaalton> https://bugs.edge.launchpad.net/~ubuntu-x-swat/+packagebugs
[07:59] <tjaalton> that's my start page when triaging
[08:01] <tjaalton> bbl
[08:01] <bryce> ah cool
[08:02] <bryce> looks like most bugs are in lrm
[08:52] <tjaalton> yeah :/
[08:53] <tjaalton> ati needs a cleanup as well
[08:53] <tjaalton> maybe I should put the BTFH hat on and start bashing :P
[09:12] <bryce> hehe :-)
[09:19] <tjaalton> bryce: did you notice my comment about the bug list you showed me? only one of them has been tested against the ~latest driver, and the rest should be fixed according to upstream (I actually closed some of them)
[09:19] <tjaalton> this was about intel vs. i810
[09:20] <bryce> oh cool, no I missed that
[09:20] <tjaalton> so it looks like i810 could be dropped IMHO
[09:20] <bryce> was that this page?  http://people.ubuntu.com/~ogasawara/qa-hardy-list-archive/sort-by-package/current-buglist.html
[09:21] <tjaalton> 12:04 < bryce> 114331  135093  147189  157610  16887  131972  and probably others
[09:21] <tjaalton> nope, that :)
[09:21] <bryce> ahh
[09:22] <tjaalton> 147189 is the one that seems to be valid now, and I forwared it upstream
[09:23] <bryce> ok
[09:24] <bryce> there were some more 8xx ones I looked at in doing the triage today.  I don't remember whether I asked for them to be re-tested or if they looked legit
[09:26] <bryce> but yeah, I think we're getting close to being ready to deprecate it
[09:26] <tjaalton> I asked Shaun (147189) to try the latest driver
[09:27] <tjaalton> since it seems that the past week there has been some changes that should fix issues with i830
[09:38] <Ng> ooh, new -intel this morning :)
[09:38] <bryce> heya Ng
[09:38] <Ng> hey hey :)
[09:39] <Ng> how evil am I for using EXA, greedy migration and "ExaNoComposite" "false"?
[09:39] <Ng> seems to make EXA as nippy as XAA
[09:40] <tjaalton> for me just the greedy stuff is enough (965)
[09:40] <tjaalton> bryce: we actually toyed with the idea to use EXA only for 965 and use greedy :)
[09:40] <tjaalton> this was on #ubuntu-desktop yesterday
[09:44] <bryce> not a bad idea actually
[09:46] <tjaalton> and XAA for the rest
[09:46] <tjaalton> I'm not sure if it's doable, but..
[09:47] <bryce> yeah, I was wondering if that's feasible myself
[09:48] <bryce> bedtime...  night
[09:48] <tjaalton> greedy makes it use CPU more, so I haven't tried what happens to the battery lifetime
[09:48] <tjaalton> night!
[09:48] <Ng> hmm
[09:48] <Ng> there's something odd about the new driver
[09:49] <tjaalton> Ng: oh? describe
[09:49] <Ng> tjaalton: it's quite hard to describe, but things like the fading in of menus is more like a flicker
[09:49] <Ng> and it almost seems like colours are made up of 8bit gif dithering
[09:50] <tjaalton> hmm
[09:51] <tjaalton> Ng: you had 945?
[09:51] <Ng> nah this is 855
[09:51] <tjaalton> ah ok
[09:53] <Ng> gonna try pulling out those options
[09:53] <Ng> it's possible my eyes have just gone funny ;)
[09:54] <Ng> no, that didn't help :/
[09:58] <Ng> going back to XAA doesn't seem to have helped either
[09:59] <tjaalton> ok, so it seems like a regression then
[10:02] <Ng> a pretty hard one to describe, but for example glxgears - the green gear looks like it alternates between light green dots and dark green dots instead of all of the pixels being a uniform green
[10:04] <Ng> I should still have the old package, I'll give that a try again just to be sure I'm not mad
[10:04] <tjaalton> hehe
[10:06] <Ng> nope, definitely not mad :)
[10:06] <Ng> 2:2.2.0+git20080107-1ubuntu2 is better
[10:07] <jcristau> should be pretty easy to bisect then
[10:07] <tjaalton> yes, since that version had all the commits up to ~Jan 31.
[10:14] <Ng> filed as #189868
[10:15] <tjaalton> thanks
[10:37] <tjaalton> looks like the only possible commit which broke this is "Allow non-strict free order for bo_list" a70b59bd44d14e77c9e522dbe225b62a8bcf3050
[10:48] <Ng> tjaalton: I'd be happy to test a build without that :)
[10:48] <jcristau> Ng: git-revert a70b59bd; make :)
[10:51] <tjaalton> Ng: I'll reply to the bug, 'patch -R < foo' should work too :)
[11:01] <tjaalton> yeah, menus work fine on my 965
[11:17] <ubotu> New bug: #189863 in linux-restricted-modules-2.6.24 (restricted) "[Hardy]Window borders are not drawn using fglrx" [Undecided,New] https://launchpad.net/bugs/189863
[11:17] <ubotu> New bug: #189868 in xserver-xorg-video-intel (main) "regression in 2.2.0.90-2ubuntu2 on i855GM" [Undecided,Incomplete] https://launchpad.net/bugs/189868
[11:44] <Ng> bah, wgetting that git link downoads 30kb of html
[11:44] <tjaalton> hmm? not here
[11:45] <Ng> in a browser I get a patch
[11:45] <tjaalton> 928 bytes
[11:46] <Ng> this is a patch to i830_memory.c, right?
[11:47] <tjaalton> yes
[11:47] <Ng> it's not reverse applying cleanly. just looking at why
[11:48] <tjaalton> it's not that huge either, so you can just edit by hand if all else fails :)
[11:50] <Ng> I sure wish patch could tell me why though
[11:53]  * Ng shrugs. visual inspections suggest I reverted it by hand fine, but the patch still doesn't apply. gonna build it and see what happens ;)
[12:00] <Ng> tjaalton: nope, still doing it without that patch, afaics
[12:00] <tjaalton> bah
[12:08] <tjaalton> Ng: look at the commits since the one 13 days ago, list found here http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-intel.git;a=shortlog;h=xf86-video-intel-2.2-branch
[12:08] <tjaalton> and apply one at a time on top of the working driver
[12:08] <tjaalton> but I guess you don't have the sources anymore?
[12:10] <tjaalton> you can find it here: http://users.tkk.fi/~tjaalton/dpkg
[12:11] <Ng> cool, ta
[13:36] <seb128> is there a known bug about middle click events being duplicates sometimes on hardy?
[15:01] <tjaalton> I don't remember seeing one
[17:15] <ubotu> New bug: #189969 in xserver-xorg-video-intel (main) "Newest revision (2.2.0.90-2ubuntu2) leaves garbage on screen" [Undecided,New] https://launchpad.net/bugs/189969
[17:16] <Ng> interesting
[17:16] <Ng> that's 855
[17:17] <Ng> it's very distressing how many variants there are of each model. I've never seen anything remotely like that bug ;/
[17:35] <tjaalton> could be the same commit that causes the breakage
[17:41] <tjaalton> Ng: even more important for you to find out which commit broke it :)
[17:48] <Ng> tjaalton: yeah, I'll try and work through the patches this weekend
[17:48] <tjaalton> hmm, I could forward it upstream in the meantime
[18:21] <tjaalton> done
[18:33] <Ng> cool, thanks
[18:36] <ubotu> New bug: #36625 in linux-restricted-modules-2.6.24 (restricted) "can't remove nvidia-glx" [Medium,Confirmed] https://launchpad.net/bugs/36625
[19:31] <ubotu> New bug: #22297 in linux-restricted-modules-2.6.22 "no icon for nvidia settings in gnome menu (dup-of: 44897)" [Undecided,New] https://launchpad.net/bugs/22297
[19:36] <tjaalton> \o/ the only remaining bug regarding i830 is fixed with 2.2.0.90
[19:36] <tjaalton> so now we only have the regression with 855 :)
[19:36] <bryce> whoohoo
[19:37] <tjaalton> so, we could drop -i810 with 2.2.1 final
[19:37] <tjaalton> I don't think it's of much value anymore, unless someone really insists on using xinerama
[19:38] <tjaalton> going through lrm bugs btw.. man, there are plenty
[19:39] <bryce> yeah
[19:40] <bryce> when I was looking at fglrx the other day, I just searched on "fglrx" so I could focus on just that subset
[19:40] <tjaalton> the good thing is that many bugs are marked against many lrm versions, so it's trivial to mark wontfix the old ones 
[19:40] <bryce> yup
[19:41] <tjaalton> bug 36625 might have the information needed to finally fix the nvidia/fglrx uninstall issues..
[19:41] <ubotu> Launchpad bug 36625 in linux-restricted-modules-2.6.24 "can't remove nvidia-glx" [Medium,Confirmed] https://launchpad.net/bugs/36625
[19:41] <tjaalton> (bumped the version, was against 2.6.17)
[19:43] <tjaalton> seems like if mesa has been updated or tried another revision of the driver, this could happen
[19:50] <ubotu> New bug: #190004 in xf86-input-evtouch (universe) "Please merge xf86-input-evdev 0.8.7-3 (universe) from debian unstable (main)" [Undecided,In progress] https://launchpad.net/bugs/190004
[19:56] <ubotu> New bug: #54335 in linux-restricted-modules-2.6.24 (restricted) "fglrx should contain install instructions" [Low,Confirmed] https://launchpad.net/bugs/54335
[19:56] <ubotu> New bug: #54966 in linux-restricted-modules-2.6.17 (restricted) "nvidia-glx boasts unnecessary dependencies" [Undecided,Won't fix] https://launchpad.net/bugs/54966
[19:58] <bryce> yeah
[19:58] <bryce> after FF I want to really hammer on some bugs - hopefully I'll have this xrandr gui done by then!
[20:00] <bryce> there's been a lot of people coming to us about keyboard and mouse problems...  makes me think we need to look into that more seriously
[20:02] <ubotu> New bug: #22255 in linux-restricted-modules-2.6.17 (restricted) "Potential deadlock when the connection to the AP is lost?" [Medium,Invalid] https://launchpad.net/bugs/22255
[20:02] <tjaalton> we should switch to input-hotplug (=evdev).. :)
[20:03] <tjaalton> ie. waiting for gravity
[20:06] <ubotu> New bug: #48276 in linux-restricted-modules-2.6.24 (restricted) "iwlist can make network tools unusable when it crash" [Medium,Incomplete] https://launchpad.net/bugs/48276
[22:46] <ubotu> New bug: #158317 in compiz (main) "Random screen flicker on Dell Inspiron 9400 (dup-of: 164589)" [Undecided,New] https://launchpad.net/bugs/158317
[22:46] <ubotu> New bug: #160747 in linux-source-2.6.20 (main) "Random black frames (screen flickering) #7.10 (dup-of: 164589)" [Undecided,New] https://launchpad.net/bugs/160747
[23:15] <ubotu> New bug: #34017 in linux-restricted-modules-2.6.24 (restricted) "Include nvidia-settings and nvidia-xconfig" [Wishlist,Fix released] https://launchpad.net/bugs/34017
[23:35] <ubotu> New bug: #39082 in linux-restricted-modules-2.6.15 (restricted) "nVidia driver in dapper API mismatch" [Medium,Invalid] https://launchpad.net/bugs/39082
[23:46] <ubotu> New bug: #126251 in linux-restricted-modules-2.6.22 "nvidia-settings menu item missing (dup-of: 44897)" [Undecided,Confirmed] https://launchpad.net/bugs/126251
[23:57] <ubotu> New bug: #149204 in linux-restricted-modules-2.6.22 "[gutsy] Xorg crashed with SIGSEGV" [Medium,New] https://launchpad.net/bugs/149204