/srv/irclogs.ubuntu.com/2009/07/10/#ubuntu-x.txt

james_wyay, new -intel, thank you01:12
brycejames_w, yep, hope it's stable.  Was looking like the number of issues fixed by updating it was greater than my interest in continuing to cherrypick ;-)01:21
james_wwell, it closed the bug that I thought was mine, so I'll be happy if that was the case01:22
james_wI'll hopefully be able to suspend next time I'm away from home :-)01:22
bryceyeah the suspend fix was the biggie01:22
bryceI was a bit aggressive at guessing on some bugs that I *think* are going to be fixed.  I figure the users will be bold in reopening the bugs if I'm wrong :-)01:23
james_wyeah, I don't think you'll have a problem with that :-)01:27
james_wI saw Jesse's post on possibly being able to capture a GPU dump when it freezes, that would be great01:28
Sarvatti made sure every bug closer i added to the dpms off fix was fixed for the original reporter by the update in x-updates, resume from suspend would be more in a kernel fix i would think unless dpms is getting turned off before suspend for some reason? there was a fix for most peoples suspend/resume problems in 2.6.31-rc2 and 2.6.30.101:33
brycejames_w, yeah I need to pull that patch and take a look at it01:34
bryceSarvatt, yeah I actually went and added several more bug#'s on top of yours.  I'm sure yours are accurate, but less confident on mine :-)01:35
james_wSarvatt: I could suspend and resume fine, but I had no X afterwards, and dpms would sometimes lead to the same thing as well01:35
Sarvatthttp://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=9e06dd39f2b6d7e35981e0d7aded618686b32ccb01:35
Sarvattah, any idea if it only happen when compiz was enabled james_w?01:36
james_wI'm happy to have the bug marked as fixed either way, as I never did what I should and invested the time to delve in to the problem01:36
james_wif it's not fixed then I will do so01:36
james_wyeah, I never tried without it actually01:36
james_w-intel has been too solid otherwise so far for me in karmic that I never thought to try without it :-)01:37
Sarvatttheres someone reporting that 965+ isnt resuming right with compiz enabled still even after the ordering fix but he hasnt been able to figure out why01:37
james_wI'm 945 for what it's worth01:38
Sarvattah good01:38
Sarvattdo you use KDE?01:38
Sarvatti think KDE locks the session and does a dpms off on suspend01:39
Sarvattseemed like most of the people having problems with it were using KDE, thats why I filed one of the bugs against acpi-support because it was using the screenblank script in there under KDE causing problems, getting rid of the screenblank script and letting just KDE handle dpms fixed it for a few people before it was fixed in the driver01:44
Sarvattthat was just for people with hangs on lid close though01:47
james_wnah, GNOME01:48
Sarvattheads up btw01:54
SarvattSetting up xserver-xorg-core (2:1.6.99.1+git20090709.e8121033-0ubuntu0sarvatt) ...01:54
Sarvattdpkg: warning: obsolete option '--print-installation-architecture', please use '--print-architecture' instead.01:54
Sarvattwhenever xserver gets updated, new dpkg change01:57
Sarvattxorg-server/debian/xsfbs/xsfbs.sh:ARCHITECTURE="$(dpkg --print-installation-architecture)"01:58
Sarvattah it'll be pulled in from debian anyway http://git.debian.org/?p=pkg-xorg/xserver/xorg-server.git;a=commit;h=7deebf983f53c505bc25171ab77fdc408f250a6e02:02
Sarvattadding patches on top of the current xserver will give the error though (not that it hurts anything)02:03
bryce:-)  <kees> "Launchpad can now import code from Git repositories." whoa02:04
keesbryce: http://blog.launchpad.net/bazaar/git-imports02:09
brycekees: bryce@chideok:/tmp$ bzr branch lp:xserver-xorg-video-intel02:11
brycebzr: ERROR: Invalid url supplied to transport: "lp:xserver-xorg-video-intel": xf86-video-intel has no default branch.02:11
brycehrm02:11
james_wwoop, seems to be fixed. Thanks all02:15
Sarvattbryce: do you have to link it first somewhere?02:20
bryceprobably02:22
brycejames_w, nice :-)02:22
Sarvatthttps://code.edge.launchpad.net/+code-imports/+new02:23
Sarvattbzr branch lp:ubuntu/karmic/mesa and libdrm work02:33
bryceSarvatt, sweet!05:33
kees~[6~[6~[6~[6~/win 1306:38
keesargh06:38
Sarvattspeak of the devil, they fixed that bug in pixman tjaalton -- http://cgit.freedesktop.org/pixman/commit/?id=0fce356762864572ae126733f657600fbb9116ce08:35
tjaaltonSarvatt: nice08:39
tjaaltonFYI, I'll be more or less offline the next two weeks..08:43
tjaaltonbryce: ^^08:43
tjaaltonhere's hoping that 1.7beta is released by then ;)08:44
brycetjaalton, okay hope it's for something fun :-)08:45
tjaaltonbryce: yeah, back to vacation. tomorrow I'll participate in a rowing contest, but just for fun08:45
tjaalton60km on a boat with seven pairs and a captain08:46
tjaaltonmy butt will be sore :/08:46
tjaaltonother than that it's just lazying at the summer cottage08:48
bryceah good, was worrying a bit you might be rowing to Ireland for some plundering ;-)08:48
tjaaltonyeah, there must be some viking blood in my veins :)08:49
Ngooh, -intel update08:59
RAOFYeah, with bonus crazy 60Hz mode.09:10
hyperaircrazy 60Hz mode?09:11
bryceRAOF, you mean return of vblank sync?09:14
RAOFbryce: No, I mean trying to drive my LVDS with a 60Hz mode that the display doesn't like one bit.10:53
RAOFThere'll be a bug filed with cool screenshots.10:53
bryceok10:57
apwSarvatt, how far away do we think the ati kms userspace support is?13:17
ScottKjames_w: It looks like dontzap is still setting xorg.conf, but that doesn't have the effect in Karmic that it did in Jaunty.  Do you know how that's supposed to be set now?13:55
james_wthrough the keyboard layout13:56
james_wso it's now per-user and doesn't require an X restart13:56
ScottKI see.13:56
ScottKtseliot: Did you plan on updating dontzap?13:56
tseliotScottK: yes, sure13:57
jcristauwhat's the use for dontzap at this point?13:57
ScottKjcristau: Making ctal-alt-backspace work.13:58
jcristauwell, no13:58
jcristauthe gnome kbd applet does that13:58
* ScottK doesn't use Gnome.13:59
jcristauthe kde kbd applet, then13:59
tseliotthe last time I tested it, setting dontzap with setxkbmap didn't work14:00
ScottKjcristau: It doesn't support that.14:00
tseliotin Karmic but it worked in Fedora14:00
jcristautseliot: so fix the server to have DontZap off by default14:00
tseliotjcristau: isn't it off by default again now?14:01
jcristauit is upstream and in fedora14:01
tseliotin the past we used what upstream decided so as to minimise maintenance14:02
ScottKtseliot: I built xserver-xorg-input-synaptics for karmic in my PPA with your patch for settling the touchpad down and it's made a huge difference with my mini 10v14:03
tseliotScottK: yes, unfortunately it can't be upstreamed as it's very specific to the mini 10v14:03
ScottKtseliot: Thanks for the patch.14:04
tseliotjames_w: does setxkbmap work for you to enable/disable zapping in Karmic? (I'm still using jaunty)14:04
james_wHaven't tried actually14:04
tseliotScottK: it's my job (OEM) ;)14:04
ScottKtseliot: Just because you got paid to do it doesn't mean I can't appreciate it.14:05
tseliottjaalton: ^^ (setxkbmap)14:05
tseliotScottK: I'm glad it helped :-)14:05
ScottKI'll also look at the KDE systemsetting module and see how hard that would be to patch instead.14:07
ScottKMaybe maco will do it since she want's to improve her C++14:07
james_wScottK: you don't need to14:08
james_was I see it14:08
ScottKHow so?14:08
james_wI could confirm if I could fathom the way through this twisty path14:08
james_wthe magic of good software design :-)14:08
james_wwhere will ./kcontrol/kxkb/kxkb_part.cpp end up?14:08
james_wor how can I access that part I mean?14:09
macodo what?14:09
ScottKmaco: We're discussing making dontzap work for KDE in Karmic14:09
macoit doesnt?14:09
james_wdoesn't work for anyone in karmic14:10
macooh. good to know, i was just about to try it14:11
macosince suddenly, today, my vga claims to have nothing connected to it. which is a lie. and it worked on wednesday14:11
macoand no, this isnt the 13 hours ago upload of -intel. im still on the version from a few days ago. and randomly deciding to ignore my VGA port after resuming from suspend isn't a first :-/14:12
tseliotI wanted to set the xkb option in xorg.conf and use setxkbmap to apply settings without having to restart X. This way it's DE agnostic14:12
jcristauthat's because there's no need for it in karmic... if the default is changed to off in the server, and kde/gnome both use libxklavier, they'll just have the option in their applet already14:13
jcristauand things will just work14:13
ScottKjcristau: OK.  So how come I don't have this option then?14:13
jcristaui don't know.  i don't use kde :)14:14
ScottKjcristau: OK.  So there is work to do and I'm trying to understand what.14:14
ScottKjames_w: That should be in systemsettings14:19
james_wfound it14:20
james_wOnly-Show-In=kde for half of the parts used by systemsettings is a bit odd14:21
tseliotI think I found it14:22
tseliotclick on  Regional & Language14:22
ScottKjames_w: KDE gets yelled at for 'duplicate' menu entries if stuff like that isn't done.14:22
ScottKtseliot: Yes?14:22
tseliotthere's a label which says "Command" and then there's a text entry which says something like setxkbmap -model pc104 -layout it,es -variant ,14:23
james_wit only shows14:23
james_win the system settings though doesn't it?14:23
james_wnot found it in my menu14:23
james_whttp://people.ubuntu.com/~jamesw/layout.png14:23
ScottKInteresting.14:24
ScottKI have that section, but it's greyed out/locked.14:24
tseliotjames_w: did you set the xkb option manually?14:24
james_wyou need to "Enable keyboard layouts" on the first tab there14:25
ScottKAh14:25
tseliotjames_w: or was the "Key sequence to kill the X server" already there?14:26
james_wI did nothing14:27
james_wjust chased down the GNOME stack and then up the KDE stack to find where the equivalent settings are14:27
tseliotnice, so, problem solved?14:27
tseliotwhen X is updated14:28
ScottKInteresting.  Works too.14:28
ScottKjames_w: Very nice.  Thanks.14:29
ScottKtseliot: Yes.  Problem solved.14:29
tseliotScottK: ok, so we can remove my patch from kdebase14:29
tseliot\o/14:29
ScottKtseliot: Yes.14:29
macokey sequence to kill the X server is in the Advanced tab14:30
tseliotjcristau: do you know if gnome has something similar in its xkb applet?14:32
* tseliot hopes so14:32
jcristautseliot: it does14:34
ScottKSo the remaining question is xfce?14:34
jcristauthat uses xklavier as well14:35
tseliotgreat, this means that we can drop dontzap (the package)14:35
tseliot\o/14:35
jcristauso it should just work, reading the options from /usr/share/X11/xkb/evdev.xml14:35
razor7hello..iḿ quite confused about ATi drivers...can anyone help? Thenka a lot!15:16
razor7hello..i'm quite confused about ATi drivers...can anyone help? Thanks a lot!15:21
razor7Hello I have an ATi x1950pro video card, but there is no more support from ATi/AMD since they say that this 2 years old card (worth u$s 200 at buytime) is no longer supported by ATi15:22
razor7In this repo I see theese two drivers:15:22
razor7    fglrx-installer - 2:8.620-0ubuntu3~jaunty15:22
razor7    xserver-xorg-video-ati - 1:6.12.2-0ubuntu1~xup~1 15:22
razor7I have added the repo and installed xserver-xorg-video-ati - 1:6.12.2-0ubuntu1~xup~1, but the program glmark just hangs wit message "Segmnentation fault"15:22
superm1tseliot, why can't that patch be upstreamed?16:09
* ScottK wonders too.16:10
ScottKWhile Mini 10v is where it really helps it seems like a generally good idea.16:10
tseliotsuperm1: because it's a workaround. A generic solution can't be found as the touchpad doesn't support multi-finger detection and reports some weird finger width/pressure values16:10
ScottKtseliot: Are there cases where the values it supresses are valid with other hardware?16:11
tseliotScottK: yes16:11
superm1tseliot, so if that's the case, why not just try to detect the specific model of this one, and only do the weird stuff on this one then?16:12
tseliotsuperm1: I thought about that too but I doubt that upstream would include it16:12
superm1tseliot, well it's not like it's some archaic wanna-be synaptics touchpad.  it's a real synaptics touchpad that just reacts a little differently than others.16:13
ScottKtseliot: I'd settle for something we could ship in our package as a quirk.16:14
superm1so i can't see why upstream wouldn't want to see it fully functional albeit with a separate code path for its quirks16:14
tseliotScottK: that could be done16:15
tseliotsuperm1: I'm not upstream ;) but I can ask them16:15
ScottKUpstream is better, but if it's something we can at least get in the archive, that'll help a lot of users.16:16
tseliothere's the bug report in upstream's bugzilla: http://bugs.freedesktop.org/show_bug.cgi?id=2161416:16
ubottuFreedesktop bug 21614 in Input/synaptics "Touchpad cursor jumps when two fingers are used" [Normal,New]16:16
tseliotmy second patch will be adopted (well, sooner or later): http://bugs.freedesktop.org/show_bug.cgi?id=2161316:17
ubottuFreedesktop bug 21613 in Input/synaptics "Property to make part of the touchpad insensitive to movements" [Enhancement,Assigned]16:17
superm1so both patches help it in different ways then, or are these two separate problems?16:21
tseliotyes16:22
superm1i suppose i worded that poorly, did you mean yes to the former or latter?16:23
* tseliot missed the "or"16:23
tseliotyes, they are separate problems16:24
superm1ah okay16:25
superm1so the first one is what would need such a case statement to flag on the broken model #16:25
tseliotyep16:26
tseliotsuperm1: if you want you can add a comment in the 1st report and ask upstream directly about adding the quirk16:29
superm1tseliot, so the thing is i think that's actually a problem on all synaptics touchpads though, is it not?16:29
superm1i'm on a studio 1340 right now, and the same thing happens, but it's not as frequent because the buttons aren't built into the pad16:30
tseliotsuperm1: no as there are different kinds of Synaptics touchpads16:30
superm1but if i touch the top right and bottom left corners of the "pad" it will happen just the same16:30
tseliotsuperm1: therefore, if the touchpad has multi-finger detection (not emulation), a different solution can be found16:31
superm1ah16:31
Sarvattjbarnes: seen this? xrandr broke in UMS somewhere between git20090630.cec9fc6f and git20090701.1e4784bf https://bugs.edge.launchpad.net/ubuntu/+source/xrandr/+bug/39449019:56
ubottuUbuntu bug 394490 in xrandr "xrandr: Configure crtc 0 invalid time" [Undecided,Confirmed]19:56
jbarnesSarvatt: hm, maybe 7e79fc?20:03
jbarnesfrom fdo bug #1957820:04
ubottuLaunchpad bug 19578 in initramfs-tools "initramfs-tools: can't install from scratch" [High,Fix released] https://launchpad.net/bugs/1957820:04
Sarvattactually it might have been something in the xserver update at the same time20:11
Sarvatthttp://cgit.freedesktop.org/xorg/xserver/commit/?h=server-1.6-branch&id=8d65439d5c950ea01ec8e1e4dd989aff0fb4c3f020:11
jbarnesah possible yeah20:11
Sarvattuploaded that on july 1st20:11
Sarvattyeah for sure it is an xserver problem, sorry about that jbarnes. one guy reverted to xserver-xorg-video-intel_2%3a2.7.99.901+git20090611.6d062e9e-0ubuntu0sarvatt~jaunty_i386.deb20:52
Sarvattand still had the problems20:52
Sarvattwhen it was working before20:52
jbarnescool20:53
jbarnesshould be reported upstream if it hasn't been already20:53
brycejbarnes, btw I finally went through all the 8xx reports that confirmed their problem on karmic alpha-2 and sent them up.  There were actually fewer than I initially estimated since some were dupes, and others were mis-tested.20:58
brycethere's another dozen tested/confirmed of 9xx bugs I'll be sending up hopefully today20:59
Sarvattif its dri support in KMS dont they need mesa 7.6?20:59
bryceand then there's another 120 which no one's replied at all, that we can probably just close :-)  -- but I'm going to give them a couple more weeks to respond21:00
jbarnesbryce: cool21:04
Sarvattdo you want us to test the newer pixman on edgers server 1.6 branch at all? or any other packages that might have problems getting updated?22:40
Sarvattyep for sure this is a problem in xserver 1.6.2 that will pop up when karmic updates to it https://bugs.edge.launchpad.net/bugs/39449022:46
ubottuUbuntu bug 394490 in xorg-server "xrandr: Configure crtc 0 invalid time" [Undecided,Confirmed]22:46
Sarvatti'm knee deep in a ton of packaging and need to get it uploaded asap and am working at the same time if anyone else wouldnt mind upstreaming it, will try to do it later tonight if not22:50
Sarvattuploading mesa on a 15kb/second cellphone connection is fun :)22:55
RAOFThere we go.  One funky scanout bug filed at bug #39802623:44
ubottuLaunchpad bug 398026 in xserver-xorg-video-intel "[GM45] Default 60Hz display mode drives LVDS incorrectly" [Undecided,New] https://launchpad.net/bugs/39802623:44
Sarvatthmmmm http://launchpadlibrarian.net/28897374/buildlog_ubuntu-karmic-amd64.xserver-xorg-video-intel_2%3A2.7.99.901%2Bgit20090710.4e4b947f-0ubuntu0sarvatt_FAILEDTOBUILD.txt.gz23:51
Sarvatthttp://cgit.freedesktop.org/xorg/xserver/commit/?id=aafcae79d17c1f802bc880d2142af7171fed75d8 ....?23:52
Sarvattcompiles fine on i386 but fails everywhere else23:53

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