[00:03] tjaalton, pst push xserver to git [05:27] bryceh: can't, need to rebase first.. and please don't push such changelog entries if you know I might have the commit :/ [05:29] but, sorry for not pushing it anyway! :) [05:30] it was an easy merge anyway, pushed now [05:50] bryceh: sorry, let me try another entré _after_ the breakfast.. [06:12] something's up with vmware - 805777 and 798830. needs rebuild? [06:16] abi hasn't changed, probably just unity fail [06:17] note that if it's using lightdm, it doesn't actually restart the xserver [06:20] Hm. It sounds not totally unlike Robert's ‘nothing draws after I log out on radeon’ bug that I'm trying to reproduce once the box finally finishes updating. [06:22] i should have vmware workstation licenses somewhere.. [06:22] I'm seriously weirded out at how few -intel bug reports there are against oneiric right now [06:22] the apport hook has to be broken or something [06:23] 3.0 is working ok, it seems.. after the catastrophic .38 that is :) [06:25] yeah most reports being filed against xorg are ending up being bugs in something else [06:52] "unable to initialize module building library", so much for trying out vmware workstation [06:53] probably multiarch related [09:18] <_asdf_> someone there? [09:18] <_asdf_> i installed the driver but i think im kinda not useing it :S [14:19] RAOF: so wait, we dont ship pipe_*.so in libegl1-mesa-drivers anymore? [14:20] also started getting a new build failure in the past few days with 7.12, do we want to ship these files? http://paste.ubuntu.com/638909/ [17:03] so is adding foreign-architecture i386, doing a apt-get update and installing wine:i386 supposed to be the way to go now for amd64? [17:04] err adding that to /etc/dpkg/dpkg.conf [18:07] I know its kinda of a long shot to ask here, but does anybody know how to make dkms packages that also installs a udev rule? [19:22] Prf_Jakob: it totally depends on how the packaging is done, for most things these days just throwing packagename.udev in debian/ is enough and you can add an override_dh_installudev line in debian/rules to adjust the priority it installs to if need be. man dh_installudev might help or looking at another package doing the same thing (virtualbox maybe?) === yofel_ is now known as yofel [19:51] Sarvatt: thanks! [22:49] Sarvatt: Correct. We don't ship pipe_*.so in libegl1-mesa-drivers because they're not built :). egl_gallium now contains *everything* [23:11] RAOF: well now gbm ships them :P [23:12] /o\ [23:14] had to add a echo "dri/usr/lib/${DEB_HOST_MULTIARCH}/gbm/*.so usr/lib/${DEB_HOST_MULTIARCH}/gbm" >> debian/libgbm1.install.in hook to the edgers packaging [23:15] Have you enabled gbm's gallium support? [23:16] i'm using origin/ubuntu packaging at the moment, trying not to have to maintain a packaging fork this time around [23:16] i'll give it 2 more weeks before it's changed enough that I have to at this rate :P [23:17] the build system has changed in fun new ways breaking the packaging at least once a week every week since UDS, been a pain in the butt :) [23:20] Fiddling 'round with configure flags is *much* more fun than trying to automake/autoconf the buildsystem :/ [23:20] RAOF: oh http://cgit.freedesktop.org/mesa/mesa/commit/?id=b18b2994eff2f51588abe29c7a2209220c9ee9c5 [23:20] guess disabling gallium would have been the better option [23:20] coming soon to a 7.11 near you :) [23:21] Heh. [23:23] thank you so much for working on the 7.11 packaging, massive amounts of changes in the span of 2 weeks right before branching made me not want to even look at it. fix it, broken next day, repeat [23:24] Physical id 0: +98.0°C (high = +86.0°C, crit = +100.0°C) [23:24] Mmmm, sweet! [23:24] shoving a 45w quad in a chassis designed for 35w dual cores makes for some sketchy mesa building :) [23:25] Although my 7.11 packaging has included a bunch of crazy mistakes :( [23:26] you saying i should look at the contents of the debs and not just blindly accept its right because it builds? :P [23:29] NO! I didn't mean to git clean -xdf! My beautiful build objects! [23:33] Ok, new plan. [23:34] Scrounge a power cable from the netbook, and use the shiny dell. [23:34] oh right I gave you a US lead, whoops [23:35] RAOF, one day I'll swap you that US lead for an aussie one [23:35] Ta :) [23:36] The US lead wouldn't be such an issue if it didn't have a ground pin. My universal connector doesn't do grounding :) [23:36] RAOF: bend it off? :P [23:44] Sarvatt: still around? [23:48] Hm, ok. The dell seems a little hardlock happy. [23:50] Prf_Jakob: kinda, it's well past beer o'clock so you'll have to forgive me if I'm not too helpful :) [23:50] RAOF: http://kernel.ubuntu.com/~sarvatt/lp761065/ [23:50] RAOF: absolutely essential on sandybridge :( [23:51] That's in the oneiric kernel, right? [23:51] there's a natty one in there [23:51] its the 2 patches that are desperately needed [23:51] they'll be in 2.6.38-11 [23:52] http://kernel.ubuntu.com/~sarvatt/lp761065/natty/ [23:52] Well, I upgraded to oneiric at the sprint. [23:52] oh crash happy on the oneiric kernel?! [23:52] Yeah. On whatever kernel I happen to be running. [23:52] you must be looking at desktop crashy, i'm using the same machine [23:52] (and haven't upgraded in a week and half) [23:53] btw if you swap a SSD in you'll probably need to pin natty's udev, oneiric udev doesn't work on this machine for me [23:53] something tells me it would if i was using a slow hdd [23:54] Sarvatt: okay, are you up for some noob packaging questions? [23:54] debian packaging* [23:54] Always :) [23:54] if not I'll go and bother somebody else. [23:55] of course, whats up? I'm a noob packager myself but other people might be able to help if I can't :) [23:55] Cool, well I just created a vmwgfx package with dkms mkdeb which was suprisingly painless, but I need to get a udev rule in there. [23:56] (udev rule is because the drm rules doesn't work for the standalone vmwgfx driver). [23:56] (which includes drm & ttm in some sort of unholy combination). [23:57] did you dkms-ify drm too? [23:57] oh that's going to be fun [23:57] we tried that for nouveau in lucid, had to patch udev and plymouth too [23:57] it just builds it into one binary so it works okay. [23:58] Oooh, hmm plymouth might be a problem, if its not running ontop of fbdev. [23:58] except for the udev rules