[00:09] RAOF, yeah, sounds plausible [00:11] plus I like one-liner patches ;-) [00:11] is that patching libxcb or xserver? [00:11] dang context-less git patches... [00:12] should be libx11 [00:12] hmm neither. [00:13] yep there it is [00:13] well, gotta run to a dentist appt. bbiab [01:13] Well, looks like I get to whip up an extra event queue for barriers. Win. [08:11] whee, bug triage for an hour and my head hasn't exploded yet [08:24] is there a difference in what multimonitor mode is used by default regarding if unity is running or not? [08:24] like, is the default cloned or extended [08:24] seems to be extended on my laptop [08:25] then again, it has .config/monitor.xml.. [08:25] meh [08:28] yeah, cloned is the default [08:29] ..but only if the second monitor is plugged in before lightdm starts [08:29] what a nice mess [11:49] cnd, hey [11:50] cnd, so gtk guys landed smooth scrolling, etc [11:50] cnd, that broke mouse wheel scrolling on Ubuntu though [11:50] cnd, mclasen says to ping some xinput guy [11:50] cnd, a log is http://pastebin.ubuntu.com/871310/ [11:50] " scroll valuator 2: vertical, increment -nan" [11:50] he says the "-nan" is the issue there [11:51] cnd, I would welcome you help debugging, that blocking the gtk update [12:18] cnd, http://pastebin.ubuntu.com/871338/ is the xinput for that mous [12:18] mouse [12:18] " increment: -nan" [12:34] rye: ok I'm here [12:34] not a huge fan of global menus, they ok when they workl [12:34] hi, people. psypher246 has installed nvidia drivers via jockey, found them not to be good and uninstalled them via jockey gui. [12:34] but they don't work so well [12:35] but the system has nvidia-curent package installed [12:35] psypher246, global menus are more likely to be discussed in #ubuntu-unity :) [12:35] i know :) [12:35] and nvidia-current left the blacklisting file so nouveau kernel module was not loadable [12:36] so the question - should uninstalling binary drivers remove nvidia-current as well? === chrisccoulson_ is now known as chrisccoulson [13:02] rye: yes, it should [13:02] psypher246, please check whether the bug with nvidia-current is already filed [13:03] actually, are you sure the driver is just disabled? [13:03] +not [13:04] unticking the box only disables the driver [13:04] and the bug should be against jockey [13:04] if there is a bug.. [13:06] tjaalton: yes all I did in jockey was de-activate the recommended driver [13:06] last time i did that and rebooted I had no issues, today i had to remove the blacklist file [13:07] psypher246: do it again and if you can reproduce it, file a bug against jockey [13:07] ok cool, will do, thanks [13:07] ("ubuntu-bug jockey") [13:07] rye: thanks for the help [13:07] psypher246, you are welcome [13:07] so, re-enable, reboot, disable, reboot [13:08] schweet [13:12] tjaalton: do you use the nouveau driver at all? [13:14] psypher246: not really [13:14] so you have never experienced this issue: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-nouveau/+bug/926714 [13:14] Launchpad bug 926714 in xserver-xorg-video-nouveau (Ubuntu) "continuous jerky mouse movements while using nouveau driver in 12.04" [Undecided,Confirmed] [13:15] I could verify it though [13:16] any attention to the issue would be great thanks. I have more issues with the prop drivers and this is the only thing bugging me on nouveau [13:17] i could pop the gf8600gt in the test box.. [13:18] tjaalton: thanks, brb rebooting after nv install [13:27] tjaalton: ok so I was not able to reproduce the jockey bug. this time when i removed the nvidia driver it deleted the blacklist file as well [13:38] psypher246: ok [13:38] psypher246: so the nouveau bug, it's about the mouse pointer stopping at times? [13:41] tjaalton: yes, not just the mouse though, then entire screen freezes, but only it seems like only when the mouse is moving, for instance, if i'm not moving the mouse and wathinvg a video, no freeze, but move the mouse while the video is playing, the mouse and the video freezes [13:47] appears to be fixed on 3.3rc6 [13:49] tjaalton: is that kernel 3.3rc6 [13:50] yes [13:50] awesome [13:50] did you experience it on earlier kernels? [13:50] haven't tried [13:51] ok cool; [13:52] you using a kernel PPA or installed it manually? [13:52] mainline ppa [13:52] now the fix needs to be found and backported to 3.2.. [14:16] fixed in 3.3rc1 to be exact, and the diff to 3.2 is 116 commits, meh.. [14:17] diff to drm/nouveau === chrisccoulson_ is now known as chrisccoulson [15:22] seb128, ok, I'll take a look today [15:22] cnd, thanks [15:22] cnd, did I give you the xinput pastebin output here? I'm not sure now [15:23] yeah [15:23] cnd, basically xinput list has " increment: -nan" [15:24] cnd, sorry it has been some hours and I rebooted since so I'm not sure how much details I gave, ping me if you need any debug infos ;-) [15:24] cnd, that's with evdev btw, but my alps pad on synaptic seems to have the same issue [15:24] seb128, I have a valid increment here on my macbook trackpad [15:24] hum [15:24] and my magic trackpad [15:25] seb128, can you get me a utouch-evemu device property file? [15:25] utouch-evemu: command not found [15:25] figure out the /dev/input/event* node [15:25] then: evemu-describe /dev/input/event* > device.prop [15:25] ok [15:26] apt-get install utouch-evemu-tools [15:28] cnd, http://pastebin.ubuntu.com/871594/ [15:29] why oh why does paste.ubuntu.com make you go through openid to get the download? [15:31] yeah, that's annoying [15:31] I guess it's to avoid sharing files in anonymous way over pastebining [15:32] seb128, I get an increment of -1.0000 here [15:32] though the device name is garbled for some reason [15:32] cnd, ok, that would be correct :p [15:32] likely a bug in utouch-evemu [15:32] cnd, do you use fedora? ;-) [15:32] so I don't really know what's going wrong for you [15:32] nope :) [15:32] cnd, it's not only me [15:32] ricotz has the same issue [15:32] hmmm [15:33] and he has it as well on edger with xorg 1.12 [15:33] seb128, can you pastebin your Xorg.0.log? [15:33] oh, I see the issue with utouch-evemu I think [15:33] pastebin downloaded with ^M instead of /n as line endings [15:34] cnd, http://people.canonical.com/~seb128/Xorg.0.log [15:36] seb128, I don't see anything wrong with the log [15:37] seb128, the next step would normally be me having to make debug builds with printf's and such [15:38] cnd, well, tell me what source need rebuild and give me a diff [15:38] but if you want to try debugging it yourself I can walk you through the steps [15:38] cnd, I can patch, build [15:38] ok [15:39] cnd, bah, it's fine for ubuntu-desktop for pitti and desrt [15:39] increment: -nan [15:39] ah, I feel not alone :p [15:39] heh [15:39] we are both on x86 [15:39] others are on amd64 [15:39] clue? [15:40] hmm, maybe [15:40] I'm on amd64 too [15:40] let me try on my netbook [15:40] aha :) [15:40] I have -nan on my i386 netbook [15:40] \o/ [15:40] seb128, I'll try to get a fix for this today [15:41] but if it slips to tomorrow, will that be a problem? [15:41] cnd, thanks a lot [15:41] cnd, no, it's just blocking the GTK 3.3.18 upload [15:41] cnd, but that can wait a day [15:42] cnd, that gtk is in the ubuntu-desktop ppa if you want to test it [15:42] they landed multitouch and smooth scrolling [15:42] (cross-posting from #ubuntu-devel hoping for more results) can someone with a nvidia (either free or proprietary) run "python /usr/share/pyshared/debtagshw/opengl.py " and tell me the renderer string please? same for the fglrx driver please :) [15:42] seb128, ok, I hope to write an upstream test case for this issue when I figure out the fix [15:43] that's what will probably be the blocker timewise [15:43] but I already have to do it for another bug [15:43] cnd, can we throw a fix in our package while you write the testcases? [15:43] cnd, well anyway please let me know when you have a patch so I can test at least locally [15:44] yeah, it's just a matter of getting work done :) [15:44] cnd, I will let you work then, thanks a lot of the quick response! [15:44] sure [15:51] mvo: that script is missing a recent unity-support-test change http://paste.ubuntu.com/871617/ [15:57] Sarvatt: awsome, thanks [15:57] Sarvatt: I fix this now [15:57] sarvatt@kyoko{~}:LIBGL_ALWAYS_SOFTWARE=1 python /usr/share/pyshared/debtagshw/opengl.py [15:57] INFO:__main__:gl vendor: VMware, Inc. [15:57] INFO:__main__:gl renderer: Gallium 0.4 on llvmpipe (LLVM 0x300) [15:57] INFO:__main__:gl version: 2.1 Mesa 8.0.1 [15:57] opengl_supported: True [16:00] mvo: [16:00] INFO:__main__:gl vendor: NVIDIA Corporation [16:00] INFO:__main__:gl renderer: GeForce GT 440/PCIe/SSE2 [16:00] INFO:__main__:gl version: 4.2.0 NVIDIA 295.20 [16:00] opengl_supported: True [16:01] thanks a lot! [16:33] hrm, hrm, this is strange, my opengl.py fails with a "BadMatch" error on fglrx, but it works on all the other ones (X_GLXMakeCurrent fails be be precise) [17:01] Sarvatt: are you guys planning on merging src:xorg for precise? or at least the cve fix? [17:02] jcristau: for sure (cve at least) [17:02] ok, cool. === yofel_ is now known as yofel [22:26] HALP HALP [22:26] latest 12.04 updates made my screen brightess set to MAX [22:35] anyone recalls where I can ECHO the bright setting? [22:41] It's in /sys somewhere, IIRC. [22:43] /sys/class/drm/$your_lvds_thingy. [22:44] Unless it's elsewhere, because laptops like to be difficult. [22:44] $ cd /proc/acpi/ [22:44] ac_adapter/ battery/ button/ [22:44] it used to be there [22:44] it's broken for a lot of people, it seems [22:44] ill file a bug [22:44] /sys/class/drm$ ll | pastebinit [22:44] http://paste.ubuntu.com/872170/ [22:45] FernandoMiguel, there are like 5 bugs no need of another duplicate [22:45] :D [22:45] seb128: anyone looking at it? [22:45] it's a gnome-settings-daemon issue [22:45] any package we can revert ? [22:45] FernandoMiguel, no, I might look at it tomorrow [22:45] thanks [22:45] g-s-d [22:47] seb128: can we revert gsd? [22:47] FernandoMiguel, you can do whatever please you it's your box [22:48] FernandoMiguel, no need to ask me for permissions ;-) [22:48] -rw-r--r-- 1 root root 461K Mar 2 07:35 gnome-settings-daemon_3.3.90-0ubuntu4_amd64.deb [22:48] seb128: just asking.... [22:48] I don't want to go blind here [22:48] just dpkg -i that [22:49] rebooting [22:50] FYI it's working again [22:51] thanks [22:51] another user on +1 just reverted to https://launchpad.net/ubuntu/+source/gnome-settings-daemon/3.3.90-0ubuntu4/+build/3253267 and is rebooting [22:59] hey I was trying to follow the instructions on https://help.ubuntu.com/community/HybridGraphics#Script_for_use_during_bootup but it is not turning my discrete GPU off and it consumes a lot of power. Can anyone help? [23:01] Huh. I'm not surprised that script doesn't work, there's no guarantee that vgaswitcheroo will be available when it runs. [23:01] eruditehermit: You should be able to turn off your discrete gpu with echo OFF | sudo tee /sys/kernel/debug/vgaswitcheroo/switch [23:02] If that doesn't work, then you've got a system that vgaswitcheroo doesn't support, and... it's difficult. [23:02] RAOF, it does work when I do it manually [23:02] Ok. [23:02] RAOF, I wanted it to be a startup script [23:04] RAOF, is there a way to make it work? [23:04] There is; let me nab you an upstart script. [23:04] ok [23:04] awesome! :) [23:08] eruditehermit: http://paste.ubuntu.com/872205/ [23:09] RAOF, what should I do with this? [23:10] drop it in /etc/init [23:10] as something.conf [23:11] thats it? [23:12] uh, i don't think you can use && and || like that in start on blocks [23:12] i think it has to be and/or [23:15] Huh. Something similar worked for me. [23:15] Maybe accidentally :) [23:17] i just checked the code - "and" and "or" are definitely the only recognized operators [23:17] so I just replace && withand [23:18] and || with or? [23:18] yeah [23:18] then I just restart [23:18] and it will do it? [23:19] you can run "sudo initctl reload-configuration; sudo initctl list | grep something" to make sure it gets picked up [23:19] (where "something" is the something in /etc/init/something.conf that you saved the file to) [23:20] i get [23:20] vgaswitcheroo stop/waiting [23:20] then yeah, you should just need to reboot [23:21] ok [23:21] will brb [23:21] thanks all [23:24] did not work [23:24] :( [23:25] echo OFF | /sys/kernel/debug/vgaswitcheroo/switch [23:25] should that be [23:25] err he meant > yeah [23:25] echo OFF > /sys/kernel/debug/vgaswitcheroo/switch [23:26] ok [23:26] will try again thanks [23:26] brb [23:28] success! [23:28] thanks RAOF, Sarvatt, broder