[00:23] <_schulte_> anyone have pointers to information on configuring lightdm?  I'm trying it out as my display manger, and I'd like to try different (e.g., webkit) themes.  thanks
[00:28] <bdrung> cjwatson: thanks
[01:30] <dpolehn> has the debian import freeze already started?
[01:32] <bryceh> _schulte_, /etc/lightdm/
[01:34] <charlie-tca> dpolehn: according the schedule, yesterday - https://wiki.ubuntu.com/OneiricReleaseSchedule
[01:34] <dpolehn> thanks
[01:35] <dpolehn> charlie-tca: so bugs need to be filed to import packages from debian correct?
[01:35] <charlie-tca> yes, I think so
[01:37] <dpolehn> thanks again
[01:41] <_schulte_> bryceh: beautiful, thanks, I should have thought to look there...
[07:40] <evfool> has someone ever used auto-upgrade-tester? can I get a bit of help with it?
[10:09] <euroford> hi all
[10:10] <euroford> i want to update gmp and gcc in lucid
[10:11] <euroford> gmp can be built without any problem
[10:11] <euroford> bug it break gcc when i upgrade gmp
[10:12] <euroford> and i just link libgmp.so.3 -> libgmp.so.10.0.2
[10:12] <euroford> gcc works again
[10:13] <euroford> and works well
[10:15] <euroford> my question is what shall i do if i want put new gmp into ppa,  to build other packages?
[10:18] <euroford> how can i install both version of gmp in pbuilder env?
[18:18] <nh2_> cnd: is there an application that displays all fingers on a touchpad as dots to show where they are?
[18:19] <cnd> nh2_, would a fingerpaint application work?
[18:19] <nh2_> cnd: I think so
[18:20] <cnd> nh2_, install qt4-demos and run /usr/lib/qt4/examples/touch/fingerpaint/fingerpaint
[18:20] <cnd> that path is off the top of my head, so you may need to fiddle with it
[18:25] <nh2_> cnd: it start a program "Finger Paint"; it does nothing, though
[18:25] <nh2_> *starts
[18:25] <cnd> nh2_, what do you mean by "it does nothing"?
[18:26] <cnd> what device are you using?
[18:26] <nh2_> cnd: Synaptics Touchpad in a Thinkpad
[18:26] <cnd> oh
[18:26] <cnd> synaptics doesn't provide full multitouch information
[18:26] <cnd> so we can get some gestures out of it
[18:26] <cnd> but you can't see the location of the touches, for example
[18:28] <nh2_> cnd: actually, I wanted to find a program to find exactly that kind of information :D (what information touchpads deliver), so I better ask you
[18:28] <cnd> ahh
[18:28] <cnd> let me think
[18:29] <cnd> I don't think we have any visual tools for semi-mt trackpads
[18:29] <nh2_> so synaptics tells you where one finger is? what does it tell for this value when you put more fingers on it?
[18:29] <cnd> all our visual tools are for fully multitouch devices
[18:29] <cnd> it gives you a bounding box of all the touches
[18:30] <cnd> unless you have a really new device, in which case it gives you the location of the first and the last touch
[18:30] <cnd> in both cases you also know how many touches there are
[18:30] <nh2_> cnd: how can I find out? I think I once had some kind of terminal app. that told me the numbers of fingers currently on the pad
[18:34] <cnd> I don't know how you can tell between the two right now; someone posted a series of patches for the new devices just last week, but I haven't had a chance to review them
[18:37] <nh2_> cnd: what is also weird is that cat and evtest don't show any output for /dev/input/event7 (which is my touchpad); both work fine on event6 though (which is the trackpoint)
[18:40] <cnd> nh2_, the synaptics X input module grabs the evdev device node
[18:40] <cnd> so you would need to change VTs and run evtest there
[18:41] <nh2_> ahaaa :) it works
[18:42] <nh2_> is there a tool for X that outputs what it grabs?
[18:42] <nh2_> or something lower-level than xev that shows me what actually happens in the hardware from the X side?
[18:45] <nh2_> cnd: and I guess synaptics does not support "quadtap"?
[18:48] <cnd> nh2_, I don't know of any tool that visualizes evdev data
[18:48] <cnd> and X does not generate any multitouch events for these devices
[18:49] <cnd> I also don't know of anything in synaptics that does something for qaudtaps :)
[19:07] <nh2_> cnd: When I use a synaptics device, should I get output from geistest?
[19:14] <cnd> nh2_, yes, you should be able to get 1-3 touch gestures, everything except rotate
[19:15] <nh2_> cnd: hmm, I don't get any output. What for example should generate output? 2-finger-move?
[19:16] <cnd> yeah
[19:16] <nh2_> hmm
[19:16] <cnd> well, geistest doesn't subscribe to one finger gestures
[19:16] <cnd> so two and three finger gestures
[19:16] <nh2_> I guess this is also the reason why touchegg doesn't wokr
[19:16] <cnd> probably :)
[19:16] <cnd> are you sure your device is multitouch?
[19:16] <cnd> it has to be semi-multitouch, not just multi-finger
[19:18] <nh2_> ugh, what is the difference between semi-mt and multi-finger?
[19:19] <nh2_> (btw scrolling up/down here in empathy works fine with 2 fingers, but I don't know if that tells anything)
[19:19] <cnd> multi-finger is when the trackpad gives one coordinate (X,Y) and says how many fingers are on the touchpad
[19:20] <cnd> semi-multitouch is when the trackpad gives two coordinates forming the bounding box of all touches, and how many touches are on the trackpad
[19:20] <cnd> btw, touches == fingers :)
[19:20] <cnd> multi-finger could provide drag gestures through uTouch, but we haven't added support
[19:21] <cnd> we tried adding it at one point, which broke people's trackpads during the natty cycle, so we reverted the change and haven't tried again yet
[19:22] <nh2_> how can I find out which type this Synaptics is?
[19:26] <nh2_> cnd: ahh. I think this is my problem: https://answers.launchpad.net/utouch/+question/157563
[19:26] <nh2_> the change you referenced to, is that already in Oneiric by now?
[19:42] <tlp> Hi. Latest alpha release/kernel will not boot on my 2007 iMac at work. Known issue? I've been using the previous kernel via grub.
[19:43] <tlp> I think there's a panic or something. Not sure if it's logged somewhere -- couldn't find it.
[19:52] <penguin42> tlp: What happens ?
[19:56] <tlp> system begins to boot as normal, then freezes. If I do the recovery mode, I see a normal dmesg boot combined with some errors (which I don't have with me unfortunately.)
[19:57] <tlp> I'm thinking not having 3.x might be screwing up the r300g video driver as well, since Unity doesn't render.
[19:59] <penguin42> bryceh: Ping
[20:00] <penguin42> bryceh: bug 803610 looks like a continuation of your fight with synaptics
[20:11] <bryceh> penguin42, you able to repro it?  you know the drill... collect a full backtrace
[20:11] <penguin42> bryceh: No I'm not, just thought it worth pointing out given the work you'd done on that one
[20:11]  * penguin42 is just triaging bugs at the moment
[20:19] <bryceh> penguin42, ah thanks
[20:20] <bryceh> penguin42, in general when you run across bugs reporting crashes without full backtraces, have them collect a full backtrace
[20:20] <bryceh> crash bugs without backtraces are about as useful as freeze bugs without gpu dumps :-)
[20:22] <penguin42> bryceh: By full do you mean a 'bt' from gdb or is the xorg.0.log ok if the debug packages are installed?
[20:23] <bryceh> 'bt full'
[20:24] <penguin42> ok
[20:24] <bryceh> Xorg.0.log never includes the state of variables and such that you get with full backtraces
[20:25] <bryceh> full backtrace also indicates what line number the failure was on
[20:25] <penguin42> sounds like it could do with triggering gdb itself if it was installed
[20:27] <bryceh> penguin42, now you understand what apport does :-)
[20:29] <bryceh> X just uses the libc backtrace routine.  It'd be nice if libc had a routine that printed gdb like full traces
[20:30] <bryceh> but for most purposes our apport hook does the trick
[20:30] <bryceh> however it isn't turned on post-release
[20:31] <penguin42> yeh, that's a shame really; it would be nice to turn it on for particularly problematic things like X
[20:32] <bryceh> it'd be ok for X crashes, those are rare enough.  but in general we'd probably be innundated with too many bug reports
[20:32] <bryceh> I mean, more than we already are :-P
[20:33] <penguin42> nod
[20:33] <bryceh> if we had a proper crash database for the post-release ones to go to, rather than filing them as bugs directly, that'd make it worth collecting them
[20:38] <bryceh> https://dev.launchpad.net/LEP/ApportOnStable