| valorie | hi folks, I was trying to see what setup my touchpad has, when I was confronted by this error message: http://wstaw.org/m/2012/10/08/plasma-desktopTN2364.png | 05:48 |
|---|---|---|
| valorie | I don't know if I should file a bug, and if so what info to include in the bug report | 05:49 |
| valorie | but something is clearly askew | 05:49 |
| tjaalton | valorie: file a bug against kde | 06:09 |
| valorie | heh | 06:12 |
| valorie | def. a KDE bug yes, but i guess I'll have to ask in the devel chan where in KDE the problem is | 06:13 |
| valorie | thanks, tjaalton | 06:13 |
| tjaalton | whatever has the capplet | 06:19 |
| tjaalton | valorie: synaptiks | 06:20 |
| valorie | ok | 06:21 |
| valorie | seems like a discovery problem | 06:21 |
| valorie | because the touchpad is working | 06:21 |
| valorie | but the "error message" is clearly insane | 06:22 |
| mlankhorst | morning | 06:41 |
| mlankhorst | interesting error indeed :-) | 06:42 |
| valorie | :-) | 06:43 |
| mlankhorst | mvo: can you look at https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1062503 ? | 07:42 |
| ubottu | Launchpad bug 1062503 in apt (Ubuntu) "apt fails to install libglapi-mesa-lts-quantal correctly on switching x stacks" [Undecided,New] | 07:42 |
| bryceh | https://bugs.launchpad.net/ubuntu/+source/xorg-server?field.status:list=NEW&field.status:list=INCOMPLETE_WITH_RESPONSE&field.status:list=CONFIRMED&field.status:list=TRIAGED&field.status:list=INPROGRESS&field.status_upstream=pending_bugwatch&field.status_upstream=hide_upstream&field.status_upstream=resolved_upstream&field.status_upstream-empty-marker=1&field.upstream_target=xorg-server&field.tag=quantal+-kubuntu+-xubuntu+-ppc+- | 08:02 |
| bryceh | omit&field.tags_combinator=ALL | 08:02 |
| bryceh | odd we have a sudden influx of xserver crashes involving __mpn_lshift() since the last few days | 08:03 |
| tjaalton | yeah I noticed the same | 08:07 |
| bryceh | probably all dupes, but didn't look into them at all | 08:09 |
| mvo | mlankhorst: I will look at it once I'm finished my current task ,thanks for the pointer | 08:10 |
| mlankhorst | yay ty :) | 08:11 |
| mlankhorst | bryceh: yeah weird bunch though, can't make sense of it, seems to be nouveau | 08:19 |
| tjaalton | not because of the patch you pushed?-) | 08:20 |
| mlankhorst | not a chance, it changes 1 constant to another | 08:21 |
| mlankhorst | seems more like a continuation of some other problem | 08:22 |
| mlankhorst | and since it's run during atexit, whatever caused it to crash already happened.. | 08:27 |
| mlankhorst | [ 23980.340] (II) NOUVEAU(0): NVEnterVT is called. | 08:28 |
| mlankhorst | [ 23980.414] (EE) NOUVEAU(0): failed to set mode: Device or resource busy | 08:28 |
| mlankhorst | Fatal server error: | 08:28 |
| mlankhorst | [ 23980.414] failed to create screen resources | 08:28 |
| mlankhorst | besides, the other one is radeon | 08:31 |
| mlankhorst | https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1061515 | 08:31 |
| ubottu | Launchpad bug 1061515 in xorg-server (Ubuntu) "Xorg crashed with SIGABRT in __mpn_lshift()" [Medium,New] | 08:31 |
| bryceh | listp=0xffffffffffffffff | 08:35 |
| bryceh | ^ do they all have that in common? | 08:36 |
| mlankhorst | didn't check, wouldn't surprise me | 08:36 |
| bryceh | #11 0x00007f3c74c3c02c in ErrorF (f=0xb42 <Address 0xb42 out of bounds>) at ../../os/log.c:817 | 08:36 |
| mlankhorst | seems exit is messed up at least | 08:37 |
| bryceh | 0xb42 out of bounds sounds suspcious too | 08:37 |
| mlankhorst | yeah it's probably overwritten | 08:37 |
| bryceh | at least 3 follow this pattern. | 08:38 |
| mlankhorst | and here I was hoping we wouldn't have a random memory corruption right before release :\ | 08:38 |
| mlankhorst | http://pastebin.com/02gPy0AY is what valgrind came up with | 08:39 |
| mlankhorst | one during xf86OutputSetEDID too, but that one is probably less harmful | 08:40 |
| bryceh | intriguing | 08:40 |
| mlankhorst | probably the same as on intel, mouse gets freed after devices are, so it tries to clean up device backing which wouldn't work.. | 08:42 |
| mlankhorst | although in this case it seems to be a refcount bug | 09:02 |
| mlankhorst | the edid one is probably harmless, but a real simple fix | 09:22 |
| mlankhorst | tjaalton: could you upload xorg-server real quick? I doubt the cursor problem will be fixed today but the other issue should be fixed | 09:34 |
| tjaalton | mlankhorst: alright | 09:36 |
| jcristau | no compat output sounds like a bug? | 09:39 |
| mlankhorst | jcristau: presumably it happens before compat_output is initialized | 09:39 |
| mlankhorst | because it happens during edid probing | 09:41 |
| jcristau | hmm | 09:41 |
| jcristau | ok.. | 09:41 |
| mlankhorst | the real annoying parts are the cursor problems though | 09:54 |
| mlankhorst | oh fun, found another one | 09:59 |
| tjaalton | hmm, might pull other fixes too | 10:00 |
| jcristau | the xserver stable branches seem to have died somehow | 10:01 |
| mlankhorst | tjaalton: yeah hitting some fun :/ | 10:02 |
| mlankhorst | http://paste.ubuntu.com/1267192/ for the acpi valgrind bug? | 10:05 |
| mlankhorst | for valgrind backtrace justifying it: http://paste.ubuntu.com/1267193/ | 10:05 |
| mlankhorst | but I need to do slightly more justification, hold on with uploading for now | 10:08 |
| mlankhorst | tjaalton: ok so the trigger for that one is killing acpid before Xorg-server is closed | 10:24 |
| tjaalton | hrm | 10:30 |
| tjaalton | why not just rip the acpi code from the server then, it's not used for anything aiui | 10:30 |
| mlankhorst | except acpi events for video probably? | 10:32 |
| mlankhorst | intel handles XF86_APM_CAPABILITY_CHANGED at least.. | 10:35 |
| mlankhorst | if it wasn't for being the last day before final freeze, I would agree that we should nuke it | 10:36 |
| jcristau | that might be replaced by drm hotplug events nowadays | 10:36 |
| jcristau | maybe... | 10:36 |
| tjaalton | riight, maybe not nuke it now :) | 10:38 |
| mlankhorst | tjaalton: ok I'm tempted to leave it for now, see if someone reviews it and if not have it in before release since it's a real crasher | 10:41 |
| mlankhorst | pushed | 10:45 |
| mvo | mlankhorst: sorry for the slow reply, I commented in the bug, would be great to get the output of the debug comment i put there | 15:23 |
| mlankhorst | ok | 15:24 |
| mlankhorst | mvo: seems to not even consider unpacking amd64 version | 15:29 |
| mlankhorst | added log :-) | 15:31 |
| === bjsnider_ is now known as bjsnider | ||
| === soren_ is now known as soren | ||
| soreau | After removing unity from my 12.04 install, X fails to display anything. It starts without any fail messages in ~/.xsession-errors and runs but doesn't display anything at all | 22:33 |
| soreau | ps ax shows X running at least but X log seems to show it exiting at the end here for no apparent reason http://pastebin.com/96zrSt4J | 22:36 |
| soreau | I tried killing X and all dm's then using startx but it exhibits the same problem | 22:37 |
| * soreau about to try with different user | 22:37 | |
| mlankhorst | soreau: vesa? | 22:41 |
| mlankhorst | not even fb0? | 22:42 |
| bryceh | soreau, check the lightdm logs | 22:50 |
| soreau | nope, not even a new user works | 22:50 |
| soreau | the screen seems to change modes but just shows illuminated black | 22:51 |
| soreau | mlankhorst: vesa is what the default is choosing | 22:51 |
| soreau | This is a really old intel gpu, I think i810 or something | 22:51 |
| mlankhorst | yeah I noticed :/ | 22:52 |
| soreau | has always worked fine until I decided to uninstall all the unity stuff instead of chmod -x | 22:52 |
| soreau | So I'm guessing somewhere in it's xinit sequence it tries to load some unity stuff and things go south from there | 22:52 |
| bryceh | jeez i810? why? | 22:53 |
| soreau | It's just an old box I use as a server really | 22:53 |
| mlankhorst | some people like pain | 22:53 |
| bryceh | heh, ok... just keep it away from the rest of us ;-) | 22:53 |
| soreau | I guess I could try starting X manually without startx, let's see | 22:54 |
| soreau | nope, same thing | 22:55 |
| mlankhorst | i can't even begin to remember how this is supposed to work to start with.. | 22:55 |
| * soreau reinstalls unity stuff as a last ditch effort | 22:55 | |
| soreau | nope, still broke | 22:56 |
| soreau | I guess it's time to reinstall. Is 12.10 out yet? | 22:57 |
| soreau | !isitout | 22:58 |
| ubottu | Quantal? Nope, it's not out. Wait some 6 months... | 22:58 |
| mlankhorst | haha | 22:58 |
| soreau | heh | 22:58 |
| mlankhorst | !forget isitout | 22:58 |
| * soreau begins preparations with a backup | 23:00 | |
| soreau | hell, it's been a good run though. Had 10.04 installed until lts-upgrading to 12.04 and of course things began to fall apart after that | 23:00 |
| soreau | so haven't reinstalled on this machine for at least a couple years now | 23:01 |
| mlankhorst | well that's good though, userspace modesetting finally dying off :p | 23:03 |
| soreau | I'm using xubuntu this time though, this unity curse is getting ridiculous and out of hand | 23:05 |
| mlankhorst | well i810 is probably below system requirements for ubuntu anywa | 23:06 |
| RAOF | You should be able to start X without unity running, though. | 23:07 |
| soreau | RAOF: X refuses to display anything. Even with startx or X :0 | 23:07 |
| soreau | tried with a new user, still nothing | 23:07 |
| RAOF | Refuses to display anything, or cleanly shuts down immediately? | 23:08 |
| mlankhorst | RAOF: https://wiki.ubuntu.com/MaartenLankhorst/DeveloperApplication ? | 23:08 |
| soreau | RAOF: ps ax show it's running but X log tells a different story http://pastebin.com/96zrSt4J | 23:08 |
| RAOF | mlankhorst: Thanks for the ping :) | 23:08 |
| RAOF | That looks like X cleanly shutdown after its last client died, which is what our X-starting tools tend to do. | 23:09 |
| soreau | well the X process never returned apparently | 23:10 |
| soreau | RAOF: and this log was taken from a 'normal' boot without doing anything extra | 23:11 |
| soreau | the power glitched this morning and when I rebooted it, I noticed it had reinstalled unity stuff. Then I removed everything unity and now it's broken | 23:11 |
| soreau | (since the next reboot after removing unity) | 23:12 |
| RAOF | Does "everything unity" include unity-greeter? | 23:12 |
| mlankhorst | night | 23:12 |
| soreau | If I try starting an X client from ssh, it seems to work but still the display will not output anything | 23:12 |
| RAOF | Because removing that'll break lightdm, because that's the greeter that we use ;) | 23:12 |
| soreau | RAOF: Looks like it | 23:13 |
| RAOF | Try installing unity-greeter; I suspect that might get lightdm displaying something, since it'll have something to display. | 23:14 |
| soreau | I heard the bongos but no display output | 23:14 |
| soreau | It's as if X is working fine but there's no output to the display | 23:15 |
| RAOF | Hm, fun. | 23:16 |
| bryceh | mesa 9 is out | 23:54 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
