/srv/irclogs.ubuntu.com/2009/02/02/#ubuntu-x.txt

=== RAOF_ is now known as RAOF__
=== RAOF__ is now known as RAOF
tjaaltondesrt: ok, that's weird..06:03
tjaaltonpwnguin: we can use dbus though, and add the other devices with a daemon like what Alexia_Death has done06:04
pwnguinoh yea06:04
desrttjaalton: any ideas?06:27
desrttjaalton: is this the sort of thing that was expected to be working out of the box?06:27
tjaaltondesrt: at some point06:57
Alexia_Deathdesrt: with my daemon it already is pretty much a matter of starting the daemon and thats it.08:07
Alexia_Deathdesrt: initial configuration is a bit of work.08:07
tseliottjaalton: it looks like my patch fixes the problem: https://bugs.launchpad.net/ubuntu/+source/xfree86-driver-synaptics/+bug/32265908:53
ubottuLaunchpad bug 322659 in xfree86-driver-synaptics "touchpad left mouse button does random other things in jaunty (dup-of: 320632)" [High,Confirmed]08:53
ubottuLaunchpad bug 320632 in xfree86-driver-synaptics "tap-to-click and edge-scrolling broken in Jaunty" [High,Confirmed]08:53
* wgrant is back on the planet now.08:53
tjaaltontseliot: synaptics 1.0.0 was released, and according to the announce those things should be autodetected now and if not, the kernel is buggy08:54
wgranttjaalton: There are no changes to that end since 0.99.308:55
tjaaltonso I'm thinking that these patches might break things for other users08:55
wgrantOnly manpage changes, FDI comments, and float properties.08:55
wgrantThe problem is that the autodetection is strange.08:55
wgrantIt decides that if it can use two-finger scrolling, it will disable edge-scrolling.08:55
tjaaltonwgrant: sure, but I meant the changes between 0.15.x -> 0.99.308:55
wgrantIt's reasonable, but a complete change in behaviour from what we have traditionally had.08:56
tjaaltonright, and that's something the kernel should tell08:56
wgrantNo.08:56
wgrantIt's policy.08:56
wgrantEverybody expects touchpads to have edge-scrolling on by default.08:56
tjaaltonKernels up to excluding 2.6.29 announce multi-touch capabilities for all08:56
tjaaltonsynaptics touchpads even if the hardware does not support it. If you have a08:56
tjaaltonsuch a touchpad, you will need to enable edge scrolling explicitly in the08:56
tjaaltonconfiguration.08:56
wgrantBut the new upstream policy is to enable it only when two-finger scrolling cannot be enabled.08:56
bryce_heya08:56
tjaaltonhi bryce_ 08:57
wgrantEvening bryce_ .08:57
tseliottjaalton: my new patch fixes what I broke with another patch therefore I think it's ok to upload it08:57
tseliothey bryce_08:57
wgranttseliot: May I see this patch?08:57
tjaaltontseliot: ah, ok :)08:57
=== bryce_ is now known as bryce-sprint
tseliotwgrant: which one?08:57
wgranttseliot: The one that isn't in the archive yet.08:57
tseliotwgrant: https://bugs.launchpad.net/ubuntu/+source/xfree86-driver-synaptics/+bug/320632/comments/5308:58
ubottuLaunchpad bug 320632 in xfree86-driver-synaptics "tap-to-click and edge-scrolling broken in Jaunty" [High,Confirmed]08:58
tjaaltonwgrant: according to the text I pasted, kernel 2.6.29 should be more wise about these issues, but maybe I'm misreading things08:58
tseliottjaalton: there's still something wrong in the detection of the touchpad edges for ALPS08:59
wgranttjaalton: It's a different, but related, issue.08:59
wgranttjaalton: Few people expect two-finger scrolling, but that's all most people will get (even with 2.6.29).08:59
tjaaltonI'm lucky not to have a touchpad, so it all is a bit irrelevant to me :P09:00
tjaaltonBMFH09:00
wgranttjaalton: Basically, pre-0.99, nothing was autodetected. Vertical edge-scrolling was always enabled, and 1-, 2- and 3-finger taps always emulated clicks on buttons 1, 2 or 3 respectively.09:01
tjaaltonwgrant: I trust your judgement, so whatever has your stamp-of-approval, I'll upload :)09:01
wgrantHeh.09:01
wgranttseliot's patches are what I would have done.09:01
tseliot;)09:01
wgrantHad I know there was going to be so much bugmail from the default changes, I would have done it myself before I went away...09:02
wgrantAnd we finally have float properties.09:02
bryce-sprinttjaalton: did you get a chance to look at the client limit patch?  I could rework that patch to s/1024/512/ if you haven't already, that should be pretty simple09:05
bryce-sprinttjaalton: the hotel wireless here is ultra slow though09:05
tjaaltonbryce-sprint: well, the proto patch uses 16bits instead of 8, so maybe 9 would do for 512, but I guess 16 doesn't hurt. the one for xorg-server is trivial09:07
tjaaltongotta run now ->09:10
wgrantHuh, DRI2 actually works.09:14
bryce-sprintwgrant: what did you need to do to flip that on?09:14
wgrantbryce-sprint: Just change the AccelMethod to UXA.09:15
wgrantIt's only a little bit quirky with some menus.09:15
bryce-sprintok good09:15
bryce-sprintwgrant: can you file bugs with the issues you find?09:16
wgrantbryce-sprint: Against what? -intel?09:17
tseliotwhen I tried UXA X started and when I launched gnome-terminal X froze (together with the kernel)09:17
wgrantThe menus do this very strange bright fadish thing when coming in and out.09:18
wgrantBut i915 with UXA and DRI2 works fine, with redirected direct rendering.09:18
wgranttseliot: Why do you go back to disabling multi-finger taps in your patch in commennt #53?09:23
wgrantDisabling on non-MacBooks, that is.09:23
tseliotwgrant: because it breaks drag and drop with physical buttons09:23
tseliotwhich is not what we want ;)09:24
wgranttseliot: You mean it does strange scrolly thing and releases early?09:26
wgrantThere must be better ways to fix that, given that I'm sure it worked before.09:26
tseliotif you have one finger on the touchpad and one to hold down the physical left button drag and drop would simply not work09:27
bryce-sprintwgrant: yes 09:27
tseliotand the cursor will just move09:28
wgranttseliot: I can drag and drop almost fine - the only problem is that it drops when I take a finger of the touchpad, rather than off the button.09:30
tseliothmm...09:41
popeybefore I file a bug report against xrandr/xorg in jaunty, I just wanted to ask if there was some fundamental change in the intel driver that might cause regressions/breakages?11:01
popey(specifically it ignoring a resolution I put either on the command line with xrandr and also if I put the correct modeline in xorg.conf)11:02
tjaaltonpopey: change since when? the driver has supported randr-1.2 since feisty, so forcing a mode needs some work11:19
popeytjaalton: well I used a little bit of gtf/xrandr magic as recently as a week or so ago, and now xorg just barfs when i try to set it to a specific res 11:22
tjaaltonpopey: hmm, that sounds odd11:23
popeytjaalton: worth filing a bug ?11:24
tjaaltonpopey: probably.. where are the errors from?11:24
popeythere is no error, the screen just goes to a different mode - uniform colour of brown, no icons, no panels11:24
popeysystem isn't locked up, i can restart gdm and get back in11:25
tjaaltonhuh, weird11:25
popeyxorg.0.log doesn't even make reference to the resolution i set in xorg.conf11:25
popeyi know it's reading the xorg.conf because i added the necessary wacom stuff and that works11:26
tjaaltonyou need to do something like here to force a mode: http://users.tkk.fi/~tjaalton/foo/xorg.conf11:26
popeyI had something similar, with a separate section for modes11:28
popeyhave modified it to be more like yours and it still ignores the res I put in11:29
popeyhttp://popey.com/~alan/xorg.conf11:30
tjaaltonlogfile might help11:30
popeyhttp://popey.com/~alan/Xorg.0.log :)11:30
tjaaltonwhat if you add 'Option "monitor-LVDS" "Generic Monitor"' to the device section?11:32
popeyoh I see, in the logs it's trying to set VGA to 1280x720?11:36
tjaaltonyes11:36
popeyok, adding that option I get a lot of blinking but no x on the panel11:37
tjaaltonanother logfile too?-)11:37
popeyhttp://popey.com/~alan/Xorg.0.log_211:38
popeyhttp://popey.com/~alan/xorg.conf_211:39
popeyoh, my bad11:39
popeyi set the option in the wrong place11:40
tjaaltondoes it work if you move it?11:41
popeyheh, no11:41
tjaaltonyeah, according to the logfile it did try 12x711:41
tjaaltonbut I've no idea why it doesn't work11:41
tjaaltonhmm, are you sure it's a 75Hz screen?11:41
popeyi used gtf to generate that modeline, and specified "1280 720 60"11:42
tjaaltonoh11:42
tjaaltonI misread it11:42
popeyworth me filing a bug do you think?11:44
tjaaltonyeah11:44
popeyalthough from the xorg log it seems to be using 1280x720 for the external VGA still and autodetecting LVDS11:45
popeyoh, then tries later11:46
popeytjaalton: further investigation reveals that actually my laptop can't switch to _any_ resolution other than the native one12:56
popeytjaalton: fyi if you're interested i filed bug 324286 about it. thanks for the help13:15
ubottuLaunchpad bug 324286 in xserver-xorg-video-intel "[945] [jaunty] blank screen on resolution change" [Undecided,New] https://launchpad.net/bugs/32428613:15
tjaaltonpopey: ok thanks13:17
=== seb128_ is now known as seb128
jcristauhrm. patches.ubuntu.com isn't getting updated?18:31
tjaaltonseems like it20:15
maxbHi, after the recent libgnome change to use the xorg detected dpi value for font sizing, my fonts are huge21:26
maxbHowever, "xrandr --verbose" is getting the correct screen dimensions21:26
maxbHardware is an Acer Aspire One21:27
maxbCan someone guide me to filing a useful bug? Thanks.21:27
jcristauhrm.21:30
jcristau    - Add libxinerama1 to build depends.21:30
jcristaureally?21:30
tjaaltonsounds odd21:38
tjaaltonmaxb: just use 'ubuntu-bug -p xserver-xorg-video-FOO', that should include the useful logs21:40
maxbtjaalton: xdpyinfo says the correct resolution and physical dimensions...does it still make sense to file against the driver?21:51
tjaaltonmaxb: the logfile will tell21:52
maxbok21:52
jcristau(probably not. but then the font size should be correct.)21:53
maxbLP 324518 filed, anyway21:55
ubottuLaunchpad bug 324518 in xserver-xorg-video-intel "Overly large fonts" [Undecided,New] https://launchpad.net/bugs/32451821:55
tjaalton(==) intel(0): DPI set to (96, 96)21:58
jcristautjaalton: i think that one's misleading21:59
jcristauit says that here, too, but xdpyinfo says 125x12521:59
tjaaltonhmm, nice21:59
superm1tjaalton, dpkg-shlibdeps complains elsewise...22:01
tjaaltonsuperm1: hum, ok22:03
jcristausuperm1: okay. i guess that makes sense. blobs ftl.22:03
tjaaltonmaxb: could be that your font sizes are just too large?-)22:03
superm1yeah. pretty much everything in that upload was working around ugly behavior that happens with the blob :(22:03
tjaaltonmaxb: I believe the defaults have been changed in the past to work around some issues, so maybe they should be changed again22:04
tjaaltonhmm, although they should be smaller now22:05
maxbtjaalton: Well, it could be that everything's working as designed, but if that's the case, then every Aspire One user is going to get their fonts blown up to an unhelpful size when upgrading22:05
tjaaltonmaxb: I'm not sure what's going on, but probably the guys at the sprint will notice it too, and do something about it :)22:05
maxbooh, good point22:13
maxbHopefully there'll be a few netbooks there22:13
maxbtjaalton: I just updated a second machine, with the same result - a distinctly excessive increase in font size. I suppose things are working as designed, but this design is liable to annoy people upgrading, I suspect22:26
tjaaltonmaxb: AIUI the effect should be the opposite, but oh well :)22:29

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