[03:04] one day i'll learn mesa master builds are broken after 7.7 branch merges almost every time... always end up packaging mesa right after it happens [03:10] Sarvatt, heya [03:16] might be a good time to update ati soon, quite alot of bug fixes in the past 2 months on master like signifigant speedups for r600 2D, low memory EXA fixes (and fixing NoAccel to work with KMS), a ton of dpms/incorrect resolution fixes, and a fix for a VT problem I've seen people having [03:16] heyo bryyce [03:17] update just the -ati ddx ? [03:17] had a question for ya, with this new archive restructuring thing, is X going to have a team or is it going to be part of desktop (or core-dev still)? [03:17] yeah [03:18] ok I'll todo getting -ati updated either this week or next [03:18] (next week is the developer sprint so I may not get much done) [03:18] re: restructuring [03:19] well there's certainly advantages either way [03:19] I think I'd like to see Ubuntu-X be its own team separate from the desktop [03:19] noone really wants to work on X stuff, really strange [03:19] heh [03:20] Sarvatt, I don't know, I guess I don't have a strong feel either way. What do you think? [03:21] it seems like "X stuff" is distinct enough from "Gtk application stuff" that it splits off easily [03:21] on the other hand, if having it be two teams means that there's an extra barrier of entry to people who might get involved, that'd be bad... [03:23] I suppose the thing to look at would be if the desktop team is set up to include both gnome and kde; if not, and if there are separate teams for each, then having X be a separate team may make sense [03:24] if gnome and kde are both included in the desktop team, then there's less reason to have X separate [03:24] probably not enough people to even make it worth splitting off, I was just reading the ubuntu-motu logs earlier and was confused where X packages would fall with the new changes since one day down the line when I have enough experience with it i'd be interested in helping out there and dont know which team or whatever I should look into joining. have to dig up the wiki pages again, was reading on my phone :) [03:27] like is there still going to be a core-dev team and the desktop team has a subset of upload permissions of that? I'm not really interested in stuff outside of X components so I've been on the fence about MOTU and it seems like thats dissolving or something [03:30] * bryyce nods [03:31] that's another reason favoring a stand-alone X team... there may be people who are *only* interested in X, and joining the full Desktop team just to get access to X stuff might seem a bit too broad for them [03:57] oh nice, didn't know -ati disabled EXA pixmaps for cards with <=32MB vram under KMS now, that should fix up those problem cards that needed XAA under UMS back in the day now that KMS is on by default [04:03] wonder why they didnt branch intel 2.10, post 2.10 is in even worse state right now than 2.10 was for me, compositing is all messed up here [04:12] bryyce, does that mean you're gonna hire more people to work on X? =] [04:27] superm1, you can bet on it [04:27] that's great! you can never have too much love hacking on X stuff === \vish is now known as vish [05:21] yay for blindly reverting random batchbuffer related commits in -intel 2.10 around when it started to try to find the problem since I cant bisect it :D [05:22] :-( [05:23] -intel master was unusable from 11-11 to 12-10 on 945 here [05:29] hmm theres a few commits that change things for just <965, maybe that'd be a better place to start [05:46] hmm, -joystick has some issues [05:46] took over the mouse [05:47] ah I had my weirdo udev rule I was experimenting with still installed [05:48] cant move the mouse away from the top gnome-panel bar with a controller plugged in [06:00] got it, adding ENV{x11_options.MapAxis1}="disable-mouse", \ to /lib/udev/rules.d/67-xorg-joystick.rules worked [06:02] hmm that cant be right, thats a MapButton option, it probably disabled that axis or something. should have read the man better [06:19] yeah every so often I have to go through doing a full reinstall in order to clear out all the experimental crap I'd been playing with :-) [06:19] which has made me be quite strict about not doing any testing on my main desktop [06:24] i've had this install since intrepid and used devel releases the day they opened since jaunty, probably really due a reinstall here sometime [06:25] ext3 converted to ext4 didn't help fsck times at all, my 40gb ext4 native partition fscks in seconds vs a minute or two for the 10gb / so I'm due a reinstall there for that too [06:25] hmm havent seen this one http://cvs.fedoraproject.org/viewvc/devel/xorg-x11-server/xserver-1.7.1-gamma-kdm-fix.patch?view=markup === seb128_ is now known as seb128 [12:58] ugh there's that segfault on USB plug again [12:58] and that wasn't even crazy confused USB afaict, that was just me unplugging my hub and plugging it back in [12:59] did I ask last time I was talking about this why apport doesn't catch Xorg server crashes? [13:01] Ng, because the xorg change to allow that was crashy [13:01] Ng, it's on the todolist for sprint to rewrite it [13:02] ah ok :) [13:02] I'd really like to be able to give a good quality bug report for this segfault other than "erm, I just kinda plug/unplug USB a bit and it crashes" :) [13:15] Ng: log in remotely and run gdb? [13:15] if it's the server crashing [13:15] tjaalton: I can't reliably reproduce it though [13:16] it's just sometimes USB insertions seem to make X segfault. weirdly it doesn't get restarted, so I'm left looking at the last state of the framebuffer and think the machine has wedged hard, but I can still ssh in [13:19] Ng: ok [15:17] Duke`_: if you want something to try -- https://edge.launchpad.net/~sarvatt/+archive/ppa [15:17] (use it on top of edgers) [15:18] Ng: there's a related bug that got fixed upstream recently iirc [15:19] Sarvatt, do I have to add it to my source.list (periodic updates) or just download it once now? [15:19] just download it once [15:19] ok [15:20] jcristau: ooh :) [15:20] what did you add in it? [15:20] i'm just blindly trying to find what broke it but its better than doing nothing for 2 more months :D [15:20] Revert "uxa: Don't treat prepare_access as a flush synchronisation point." [15:20] Commit 637f003b047e426901cab6b1fe3a0924bcb9a38a [15:21] is it compatible amd64? I'm running X86_64 [15:21] uptime [15:21] 10:21:05 up 10:40, 3 users, load average: 0.02, 0.05, 0.15 [15:21] ah, build finished. [15:21] yeah whenever it builds, i just uploaded it [15:22] whoa [15:22] mozilla must be on a break :D [15:22] Ng: http://bugs.freedesktop.org/show_bug.cgi?id=25640 [15:22] Freedesktop bug 25640 in Server/general "Reattaching USB keyboard causes double free" [Critical,Assigned] [15:23] ah, that was pointed out to me last time I mentioned this. The one key difference is that theirs was really reliably reproducible [15:23] hmm i thought that was fixed a month ago [15:23] mine really isn't [15:24] I plug in to a USB hub at least 5 times a week and it doesn't even crash once a week atm [15:24] Sarvatt: the patch is 2 weeks old, but not pushed, i think [15:28] ah thats a totally different issue now that i'm looking at it, it was Xi: reset device properties to NULL after deleting them. (#25374) that I was remembering === yofel_ is now known as yofel [16:39] hmmm... suspend does not seem to work anymore with nouveau [16:39] suspending with nvidia on the same kernel works (but gives graphical artifacts on resume) === wmerriam_ is now known as merriam [18:37] bryyce, hey ... how did that testing go? === rickspencer3_ is now known as rickspencer3 [19:07] apw, sorry, got sidetracked on another issue. It's on the todo list for today. [19:08] bryyce, np [19:57] jcristau: say, can this bug get some love? I'd like to figure out what next-steps are: https://bugs.freedesktop.org/show_bug.cgi?id=9209 [19:57] Freedesktop bug 9209 in Server/general "Xlib: Maximum number of clients reached" [Normal,New] [19:58] talk to ajax, i guess [19:58] hm, ok [19:58] * kees when is he usually online? [19:58] oh, he is, just not in here [19:59] he's at RH in boston, so daytime there === woozy_ is now known as woozy [22:13] hi [22:13] anyone to help me configuring my display settings with Dell Latitude E5400 / Ubuntu here ? [22:15] AssertFalse: also see /topic :) [22:15] lots of good links there [22:17] have tryied many many thing involving xrandr, modeline, xorg.conf but I can't get my resolution above 1024x800 ; which (i guess) is not the max I can get [22:18] what resolution you can get depends mostly on the monitor.. [22:18] (and the /topic is mysteriously not visible with irssi) [22:18] https://wiki.ubuntu.com/X [22:18] that's /topic [22:19] xrandr says "Screen 0: minimum 320 x 200, current 1280 x 800, maximum 8192 x 8192", am I misunderstanding that i can get 8192 x 8192 resolution ? [22:19] probably not [22:20] it's the max size of the framebuffer. but your monitors won't display that much [22:20] i guess [22:23] Hi [22:23] I've set a wrong resolution on my laptop , how do i reset it to default in a TTY ? [22:24] X has detected wrong refresh rates, i'm getting a black screen on lower resolutions [22:29] strycore, try "DISPLAY=:0 xrandr --output --mode " [22:29] where you get the output and mode names from "DISPLAY=:0 xrandr" [22:32] arrh , I get a X Error of failed request: Badmatch (invalid parameters or attributes) [22:33] Major opcode : 148 (RANDR) , Minor: 7 (RRSetScreenSize) [22:33] if that makes any sense [22:38] When you run “DISPLAY=:0 xrandr”? [22:40] i got the wrong resolution with gnome-display-properties [22:40] RAOF, no with the output and name [22:40] DISPLAY=:0 xrandr --output LVDS --mode 1280x800 [22:41] And “DISPLAY=:0 xrandr” had LVDS as an output and 1280x800 as a possible mode? [22:41] yes that's the native resolution [22:41] Can you pastebin the output of “DISPLAY=:0 xrandr -q”? [22:42] I'm afraid i can't , I would need a command line version of pastebin ;) [22:42] pastebinit! [22:42] sudo aptitude install pastebinit. [22:42] The worlds finest remote debugging tool. [22:43] ah ok :) [22:43] Then you'd run “DISPLAY=:0 xrandr -q | pasetbinit” [22:44] hmm, I would need network-manager now :( [22:44] oh wait, xterm session works [22:45] Oh. You don't have your wireless set up as a system-wide thing? That's a good thing to do :) [22:46] I know but I haven't configured my laptop that much since i installed lucid [22:48] Heh. Setting my home's wireless to system-wide is one of the first things I do - for pretty much this reason :) [22:49] well , I'm trying to figure out in gnome-display-properties' source code where it saves its settings [22:51] In ~/.config/gnome-something-or-other/monitors.xml [22:52] ah [22:53] i typed Alt+F2 and typed xrandr -s 0 blindly , it worked :) [22:54] :) [22:54] how does xrandr "guesses" the resolutions and refresh rate ? [22:55] 1360x768 , that pretty new to me ! [22:55] By asking your monitor for the resolutions and refresh rates it supports. [22:56] IE: X grabs the EDID via DDC. [22:57] Whoops! Phoronix is wrong about xserver-xorg-video-nouveau. [22:57] Also, yay for the kernel team! [22:57] http://pastebin.com/m7b1ec797 [22:58] 1280x800 and 1024x768 works , that's all [22:59] None of the other ones do? [23:00] I'm surprised by that output, though. [23:00] nope, black screen [23:00] You might want to submit a bug against your video driver; it would appear to be returning invalid modes (which probably means that your monitor has a broken EDID). [23:01] I'm pretty sure there's something wrong with the monitor [23:02] And filing a bug with the EDID attached means we (or, rather, upstream) can add a work-around. [23:02] I had a really hard time to get it working on gutsy and dapper (back when fglrx supported my chipset) [23:05] https://wiki.ubuntu.com/X/Troubleshooting/Resolution is the page of wonder. [23:13] https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/513011 [23:13] Ubuntu bug 513011 in xorg "Wrong refresh rates on low resolutions with HP NX 7000 on Lucid" [Undecided,New]