[00:35] <bryce_> tjaalton: awesome
[00:44] <jordi1983> I have an idea, and I want to know do you all think, wouldn't it be cool to have a PPA containing daily builds for the last ati r6xx/r7xx 3D support in order to be tested on ubuntu, and this way helping to speed up their stabilization and development?
[00:51] <jordi1983> so, I'm going to re-send the previous message, for you all to be able to read it
[00:52] <jordi1983> there ( http://www.x.org/wiki/radeonhd%3Aexperimental_3D)  it explains how to build and install the files but I'd like somebody to do the packaging in a PPA I don't know how to do it, I think there is a lot of dependencies and things you need to know to package a video driver in the right way
[00:54] <ripps> jordi1983: xorg-edgers and tormod do this already
[00:54] <Sarvatt> the drm is already done  in xorg edgers and tormod builds radeon-rewrite just about every day too..
[00:54] <ripps> hivemind
[00:54] <jordi1983> the radeon-rewrite branch applies only for r1xxx-r5xx cards
[00:54] <Sarvatt> r600-700 drm hasnt even been updated in almost 2 months
[00:55] <jordi1983> I'm talking about r6xx-r7xx-3d branch, or r6xx-rewrite
[00:57] <Sarvatt> http://cgit.freedesktop.org/mesa/mesa/log/?h=r6xx-r7xx-support
[00:57] <Sarvatt> that hardly justifies a daily build there :D
[00:58] <jordi1983> the code in that branch provides drm support for that cards and therefore EXA and Xvideo, but I'm talking about the experimental 3D support
[01:00] <jordi1983> I mean DRI (or DRI2 I don't know which one it's been implemented now)
[01:00] <Sarvatt> (i linked mesa/mesa not mesa/drm)
[01:00] <jordi1983> oops, sorry
[01:02] <jordi1983> look at this one: http://cgit.freedesktop.org/~agd5f/drm/log/?h=r6xx-r7xx-3d
[01:04] <jordi1983> on the xorg wiki says: 3D: experimental, development in two places simultaneously; mesa r6xx-r7xx-support, r6xx-rewrite, both need drm r6xx-r7xx-3d
[01:06] <jordi1983> looking at the development pace maybe daily builds wouldn't be very useful, maybe weekly builds, but something at least, right there isn't any builds at all
[01:06] <jordi1983> right now...
[01:13] <jordi1983> I think you need to have a git version of radeon, mesa and drm to test it
[01:13] <Sarvatt> tormod might be up for it, but hes on vacation for the weekend. maybe send him a message through launchpad?
[01:17] <jordi1983> Ok, I'll do it, thank you. Meanwhile, do you (or somebody else) know where can I learn how to do it myself? My knowledge about these technologies is very poor and about packaging too.
[01:18] <jordi1983> Is there any book or website or wiki...
[01:30] <jordi1983> no answer, am I writing in the wrong place?
[01:54] <Sarvatt> i was testing merging everything to see if it'd work for xorg-edgers but no luck, r6xx-rewrite is still based on a really old mesa version. maybe once they start merging with master like radeon-rewrite is we can do it on the mesa end
[01:58] <Sarvatt> cant think of a single place i learned it outside of google and apt-get source-ing stuff to see how it works in the debian folder, sorry i dont have any links handy..
[01:58] <Sarvatt> eww r6xx-r7xx support branch is based off of mesa 7.2 even
[02:33] <Sarvatt> anyone had any experience with radeon-rewrite? thinking about merging it with the karmic mesa 7.6 on xorg-edgers
[08:58] <maxb> Whoa, weird. The 2.7.1 intel driver has apparently done something weird with the aspect ratio of X elements
[08:59] <maxb> It now thinks I'm 800x600, when in fact I'm 1024x600
[09:01] <Sarvatt> tjaalton: you forgot to add 01_gen_pci_ids.diff  to xserver-xorg-video-intel, debian dropped it
[09:01] <Sarvatt> people are getting fallbacks to vesa because of it
[09:06] <Sarvatt> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/379504
[11:03] <hyperair> 9011 objects
[11:03] <hyperair> -2062495744 object bytes
[11:03] <hyperair> how fun.
[11:04] <hyperair> it's leaked so damn much memory, it even overflowed.
[11:04]  * hyperair facepalms
[18:25] <tjaalton> Sarvatt: I didn't forget anything, because I didn't know it was needed anymore
[18:25] <tjaalton> strange that it has gone unnoticed in debian
[18:33] <jcristau> hmm.
[18:35] <tjaalton> so the autoconfiguration logic is flawed, the internal table is only used if there's no config file
[18:35] <tjaalton> but that was and is well known
[18:35] <tjaalton> so maybe it was just an oversight?
[18:36] <jcristau> the internal table is used here..
[18:36] <tjaalton> let me try it myself
[18:36] <jcristau> julien@radis:~$ grep intel /etc/X11/xorg.conf
[18:36] <jcristau> julien@radis:~$ grep intel /var/log/Xorg.0.log|head -n 1
[18:36] <jcristau> (==) Matched intel for the autoconfigured driver
[18:36] <tjaalton> just dist-upgraded
[18:40] <jcristau> might be broken when /usr/share/xserver-xorg/pci/ exists but doesn't have the right entry
[18:40] <tjaalton> doesn't work here. is the newer xorg-server needed?
[18:41] <tjaalton> ah
[18:42] <jcristau> (==) Registering 'vesa' as fallback << where does this message come from?
[18:43] <jcristau> aha
[18:43] <jcristau> debian/patches/143_default_to_vesa.patch:+        xf86Msg(X_DEFAULT, "Registerin
[18:43] <tjaalton> hehe
[18:43] <jcristau> bad ubuntu patch ;)
[18:43] <tjaalton> anyway, confirmed
[18:44] <tjaalton> purging other drivers made it work again
[19:49] <tjaalton> nice, libaudit-dev is in universe, and needs to be moved to main before the new xorg-server builds
[19:49] <jcristau> tjaalton: if that's a problem you can build without selinux i guess
[19:50] <tjaalton> jcristau: yeah, probably will do that for now
[20:47] <hyperair> huh this is strange. my RAM is being pushed into the swap, but neither cache nor actual RAM usage seems to be high enough to push it into swap
[20:47] <hyperair> Mem:          1977       1580        397          0         16        516
[20:47]  * hyperair suspects GEM to be guilty.
[21:30] <hyperair> oh cool. no longer unfreeable cache, but unswapoffable free memory.
[22:21] <hyperair> Sarvatt: are there any plans to backport xorg-server to jaunty in the ppa?