/srv/irclogs.ubuntu.com/2011/10/27/#ubuntu-x.txt

=== kinoucho` is now known as kinouchou
=== yofel_ is now known as yofel
Sarvattricotz: sorry about that, missed your libjpeg-dev build-dep commit15:07
ricotzSarvatt, dont worry15:25
ricotzSarvatt, jfyi, the jpeg patch wont work :\15:35
Sarvattricotz: so has wayland from edgers ever worked for you?15:44
Sarvattricotz: http://sarvatt.com/downloads/wayland.png that was the first one I uploaded, still the same today15:45
Sarvattoh its using egl_gallium.so, no wonder15:47
LLStarksmorning sarvatt15:48
SarvattEGL_DRIVER=egl_dri2 wstart is a *little* better, background is still screwed up the same way but the terminal comes up15:50
Sarvatthiya15:50
LLStarkswayland?15:51
LLStarkswish there was an easier way to use it aside from the demos15:52
Sarvatthttp://sarvatt.com/downloads/wayland2.png looks lovely :)15:57
LLStarksyeah, getting that with edgers15:59
LLStarksprecise is fine15:59
SarvattLLStarks: so what kind of touchpad do you have that is screwed up in edgers?16:01
LLStarkssynaptics16:02
LLStarksedgers oneiric driver has hypersensitive defaults16:02
LLStarksregular natty/oneiric/precise picks up my finger hovering over the touchpad16:03
Sarvatthmm16:07
Sarvattdoes synclient ResolutionDetect=0 change it any?16:08
Sarvattdistro specific patch that might interfere with upstream changes16:08
Sarvattits crazy i dont have any machines with synaptics touchpads anymore16:10
Sarvatteverything is alps now16:10
Sarvattguess i could try synaptics on a mac's bcm597416:11
LLStarksthat synclient command doesn't help. i can mess around with gpointing settings, but i have no idea what the options do16:21
LLStarksand they sometimes help, but not enough to make usable16:21
ricotzSarvatt, i dont have it running well here neither16:58
bjsniderricotz, re: the gold linker, i talked to the developer and he says there was a patch in july that fixed the kernel issue, which is why your precise build worked. that patch would also be in oneiric's binutils-gold, so there are no worries int he present or future17:05
ricotzbjsnider, alright, maybe this fix is worth backporting to lucid then17:16
bjsniderit is17:19
bjsniderbut, backport the patch to binutils-gold or backport an update to everything that uses dkms to force the normal linker?17:27
ricotzbjsnider, backporting the binutils fix17:41
bjsnideri'll check to see if it builds17:43
ricotzbjsnider, to be clear i meant like backporting this specific fix to 2.20.117:45
bjsnideryep, i understood17:52
bjsniderthe questionis whether it will build against a 2 year old version, but i will check it out17:53
bjsniderricotz, found the patch, tested it against the binutils package in lucid, and it works. it is a relatively small change. so what is the next step?18:28
ricotzbjsnider, i think link the bug against binutils attach a debdiff and ping doko to explain it18:44
ricotzbjsnider, #ubuntu-devel18:45
bjsnidercool18:45
ricotzbjsnider, you have checked the dmks build to confirm it works?18:47
ricotz*dkms18:47
bjsniderno18:48
bjsnideri'd need a guinea pig using lucid for that19:01
ricotzbjsnider, i guess if you get it into lucid-proposed you will find one19:13
Sarvattzomg, is that what I think it is? http://bazaar.launchpad.net/~unity-team/unity/trunk/revision/1728#plugins/unityshell/src/GeisAdapter.cpp22:41
Sarvattunity working with edgers? say it isn't so!22:41
* Sarvatt prods the macbook air to build faster22:42
Sarvattsure as heck looks like it'd fix https://bugs.launchpad.net/ubuntu/+source/unity/+bug/86070722:43
ubot4Launchpad bug 860707 in unity (Ubuntu) (and 1 other project) "Unity crashes when started in an environment without utouch support (affects: 41) (dups: 2) (heat: 194)" [High,Confirmed]22:43
Sarvattawesome, it does - https://bugs.launchpad.net/ubuntu/+source/unity/+bug/85062322:47
ubot4Sarvatt: Error: Bug #850623 not found.22:47
brycehnice22:47
Sarvattbeen a real PITA not being able to use edgers directly for the past month and a half :)22:47
RAOFI have the multitouch server patches half-updated here.  Now that it seems like there's a workaround for the fn-f7 thingy I might finish that before kernel diving.22:54
Sarvattguess i should package up xf86-input-mtrack properly since its pretty much required on this MBA22:55
SarvattRAOF: -background none from lightdm is just racy, i get it maybe 2/10 boots on this machine22:56
* Sarvatt isn't surprised22:56
RAOFThis is a fast sandybridge?  What happens on the failed boots?22:57
Sarvattjust no -background none, I dont even see the splash because it boots so fast22:57
Sarvattif X is started in <8 seconds or so I dont get it22:57
Sarvattunity doesn't build in parallel? ugh!22:58
RAOFDoes here?23:00
RAOFAlthough cmake *does* manage the difficult feat of being a more annoying build system than autotools.23:01
Sarvattdoes it?23:03
SarvattDEB_BUILD_OPTIONS="parallel=4" debuild -uc -us -b sure as heck doesn't23:04
Sarvattwatching it bounce around 100% cpu usage between the cores23:04
Sarvattoh wow, xf86-input-mtrack is in debian now23:12
RAOFWhy do we have another non-evdev input DDX, and why do Airs need it?23:18
brycehwould be sad if it's perceived as easier to create a whole driver from scratch than it is to update the existing driver23:23
Sarvatti spent a few hours trying to make synaptics sane and gave up when this worked perfectly out of the box :)23:29
RAOFTo some extent it will be; the kernel has a crazy lead-time for driver improvements.23:29
Sarvattno luck with the unity change, avoids GeisAdapter, still crashes in libutouch-geis23:30
Sarvattunity-2d it is23:30
Sarvatttime to rebuild unity with debug symbols23:40
RAOFI suggest also DEB_BUILD_OPTIONS="noopt nostrip"23:48
RAOFOrlando: 27C, Haze.23:49
RAOFYeah, that sounds about right.23:49
SarvattInside the conference rooms: 10C23:51

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!