=== yofel_ is now known as yofel
[11:48] <RAOF> Sarvatt: You here, dude?  Have you seen terrible performance on nvidia *and* nouveau recently?  Like really horrible cairo rendering speed and slowish text rendering?
[12:58] <Vi0L0> hi, which xserver version will 10.10 use? :)
[12:58] <Vi0L0> @ www i can see that 10.10 is now using xserver 1.8, but maybe there are chances for xserver 1.9? :)
[14:37] <ginggs> hi - I'm looking for someone to open the lucid task on the following bug please?
[14:37] <ginggs> https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/553415
[14:37] <ubot4> Launchpad bug 553415 in xorg-server (Ubuntu) (and 2 other projects) "mouse trapped in box for Open Motif (affects: 24) (dups: 3) (heat: 107)" [Undecided,Fix released]
[14:42] <jcristau> why am i not surprised how much of a mess that bug is..
[15:02] <jcristau> RAOF: ginggs was asking about bug 553415 for lucid, which apparently needs e7154e9375e6b624db01a787d9ec6c8cedc2eb81 from 1.7.6.901 cherry-picked
[15:02] <ubot4> Launchpad bug 553415 in xorg-server (Ubuntu) (and 2 other projects) "mouse trapped in box for Open Motif (affects: 24) (dups: 3) (heat: 107)" [Undecided,Fix released] https://launchpad.net/bugs/553415
[15:04] <RAOF> jcristau: Thanks.
[15:08] <RAOF> Hm.  I think it actually wants 1c612acca8568fcdf9761d23f112adaf4d496f1b to be cherry-picked, as the original patch got reverted.
[15:09] <jcristau> e7154e9375e6b624db01a787d9ec6c8cedc2eb81 is 1.7-branch's cherry-pick of 1c612acca8568fcdf9761d23f112adaf4d496f1b :)
[15:10] <RAOF> Aah, ok.
[15:19] <ginggs> I have another request too, if I may: https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers/+bug/589250
[15:19] <ubot4> Launchpad bug 589250 in nvidia-graphics-drivers (Ubuntu) (and 1 other project) "Unable to resolve function glX... (affects: 2) (heat: 86)" [Undecided,New]
[15:19] <ginggs> relating to the removal of /usr/lib/libGL.so.1 in lucid
[15:20] <jcristau> that's not a request..
[15:20] <ginggs> I work at a university and spend many hours trying to get commercial packages to work in Ubuntu
[15:21] <jcristau> that's a bug in the software you're using
[15:21] <ginggs> well my request is how to proceed with this
[15:21] <RAOF> You can work around it by providing that symlink.
[15:23] <RAOF> We can't provide that symlink without breaking other stuff.
[15:24] <ginggs> yes, i realize the problem is the path is hard-coded in the application, but these are the kind of vendors whose standard answer is use RHEL or SLED or don't talk to us.
[15:24] <ginggs> you have answered my question, thanks
[15:25] <RAOF> It will be safe for you to add that symlink.
[16:16] <ginggs> thanks RAOF and jcristau, i'm out of here
[16:31] <Sarvatt> RAOF: nope, what kind of environment are you seeing it on? lucid+nouveau+xorg-edgers is fine
[16:37] <ricotz> RAOF, Sarvatt, i think i can confirm this cairo slowness using docky with maverick+nouveau+xorg-edgers
[16:41] <Sarvatt> how about without docky and without compositing?
[16:42] <Sarvatt> oh i have metacity compositing on there
[16:45] <ricotz> Sarvatt, what are you using to test the cairo "speed"? I just notice some bumpy transitions in docky
[16:45] <Sarvatt> everyday use
[16:46] <Sarvatt> my wife uses that laptop all the time and i'm always screwing with it, would have noticed it slow down
[16:47] <ricotz> it isnt noticable in everyday use for me
[16:47] <ricotz> inkscape seems to suffer a lot
[16:48] <Sarvatt> RAOF: what's slow?
[16:49] <Sarvatt> i was assuming he meant firefox and gtk stuff
[16:50] <ricotz> Sarvatt, perhaps he meant gnome-do
[16:51] <ricotz> i can confirm DO it is very slow
[16:51] <seb128> ricotz, hey
[16:52] <seb128> ricotz, didn't you have gnome-shell changes in your ppa to workaround libmozjs issues?
[16:52] <seb128> ricotz, I've diffed the rules with maverick but it seems it has no workaround
[16:53] <ricotz> seb128, hi, havent i gave you the patch in a pastebin
[16:53] <ricotz> seb128, the patch is included in the tarball to workarround the need of an autoreconf
[16:53] <seb128> ricotz, right, but I was busy and I didn't note it, and I'm wondering how it built in your ppa without those changes
[16:54] <seb128> ricotz, oh ok
[16:54] <seb128> I though it was a debian/rules change
[16:54] <seb128> thanks
[16:55] <ricotz> seb128, have you looked into gdkpixbuf2.0 and gtk+3.0 packaging?
[16:56] <seb128> sort of
[16:56] <ricotz> ok
[16:58] <ricotz> seb128, what version of g-s do you want to package?
[16:59] <ricotz> seb128, only 2.31.4 will work, 2.31.5 needs gtk3 and mutter compiled against gtk3
[17:08] <RAOF> Sarvatt: Summon gnome-do, type anything.  Watch the animation take > 5 seconds before Do is responsive again.
[17:08] <RAOF> Sarvatt: Also, text rendering in gnome-terminal.
[17:09] <seb128> ricotz, what debian has, it's a direct sync
[17:09] <RAOF> Sarvatt: maverick / {xorg-edgers,main archive} / 9300 GS
[17:09] <seb128> ricotz, gdk-pixbuf
[17:10] <seb128> ricotz, it's in the pkg-gnome svn
[17:10] <seb128> work in progress though
[17:10] <seb128> it needs extra changes
[17:10] <RAOF> Sarvatt: Aaaaand it's time for beer.  Sorry!
[17:11] <ricotz> seb128, can you give me a direct link to this gdk-pixbuf svn?
[17:11] <seb128> robert_ancell is working on update gtk now
[17:11] <Sarvatt> gnome-terminal is fine here, sounds like gnome-do is doing something bad with cairo
[17:11] <seb128> ricotz, http://pkg-gnome.alioth.debian.org/svn.html
[17:12] <ricotz> seb128, thanks
[17:12] <seb128> ricotz, it's in desktop, experimental, gdk-pixbuf
[17:12] <Sarvatt> well does it only happen with nvidia machines + gnome-do?
[17:12] <seb128> http://svn.debian.org/viewsvn/pkg-gnome/desktop/experimental/gdk-pixbuf/
[17:12] <seb128> ricotz, ^
[17:12]  * Sarvatt doesn't use it
[17:13] <ricotz> Sarvatt, might be the case, so it could be a nouveau/mesa problem
[17:13] <ricotz> seb128, thanks, i will have a look
[17:14] <Sarvatt> metacity compositing would rule out mesa at least, i can't steal the machine from the wife at the moment to test it :)
[17:16] <Sarvatt> trying it out on intel now, lessee
[17:16] <ricotz> Sarvatt, ok
[17:16] <Sarvatt> instant on intel+edgers+maverick
[17:17] <Sarvatt> any of you guys gotten a profile of it?
[17:29] <Sarvatt> go figure oprofile is busted because of binutils yet again :)
[17:30] <Sarvatt> it is spending most of its time in cairo though
[18:47] <Sarvatt> crap updated that machine to the ia32-libs in edgers, is that still broken with flash?
[18:49] <ricotz> Sarvatt, they should work 
[18:50] <ricotz> are you going to update the mesa packaging to provide this "dri-experimental" package?
[18:50] <Sarvatt> ah ok, wife's gonna complain otherwise :) btw if you want a sysprof that works to see where its spending its time on nouveau i've got one in here - https://edge.launchpad.net/~sarvatt/+archive/ppa
[18:50] <Sarvatt> one day.. probably not until 7.9 branches
[18:51] <ricotz> Sarvatt, flash stuff is working here without problems, didnt hear any comlains lately :P
[18:52] <ricotz> complains*
[18:52] <ricotz> ok
[18:52] <Sarvatt> yeah it was lucid that had the problem though, i've got that machine on lucid so she doesn't have to go through the plymouth hell we're going through again because that put her off ubuntu last time :)
[18:53] <ricotz> ok, so you might want to test it while you already installed it ;)
[19:39] <Sarvatt> yay i love when 003_recognize_glibc_2.3.2_locale_names.diff in libx11 needs refreshing :)
[19:39] <Sarvatt> any specific gnome-do plugins that you guys are using over stock to make it slow maybe?
[19:41] <jcristau> Sarvatt: if somebody upstreams that patch i'll owe him a lot of beer
[19:43] <Sarvatt> did some of it already get upstream? it seems a lot smaller than it used to be
[20:01] <Sarvatt> luckily it was just the cvs tag purge, just had to remove 2 lines and adjust the first hunk
[20:01] <Sarvatt> ricotz: any specific gnome-do plugins that you guys are using over stock to make it slow maybe?
[20:47] <ricotz> Sarvatt, it is not related to any plugin, you can literally see the drawing of the svg icons