| lesshaste | bryce: are you about by any chance? | 10:58 |
|---|---|---|
| Ng | lesshaste: it's still very early in the united states of bryce | 11:00 |
| lesshaste | Ng: ah yes. Thanks | 11:00 |
| mvo | hm, my X did restart in the middle of a jaunty->karmic upgrade | 14:04 |
| tjaalton | that's bad :) | 14:08 |
| tjaalton | real hw or virtual? | 14:08 |
| Ng | isn't that usually gdm at work? I had that a few times while scott was changing upstart gdm stuff | 14:09 |
| tjaalton | could be | 14:09 |
| apw | anyone say what X needs to connect to acpid at boot for? | 14:15 |
| jcristau | not much, iirc fedora disables that | 14:15 |
| tjaalton | it doesn't | 14:16 |
| tjaalton | right, it _does_ but shouldn't need to | 14:16 |
| apw | so it failing to connect should be non-fatal and uninteresting then? | 14:17 |
| apw | as it seems the new faster boot is triggering it to fail on some systems, cause acpid is not yet up | 14:17 |
| tjaalton | http://cvs.fedoraproject.org/viewvc/devel/xorg-x11-server/xserver-1.6.0-less-acpi-brokenness.patch?revision=1.1&view=markup | 14:17 |
| jcristau | apw: yes, it's non-fatal and uninteresting | 14:18 |
| apw | most exceelent ... thanks for the info | 14:18 |
| tjaalton | right | 14:18 |
| tjaalton | probably worth getting that patch, and also not trying i810 on intel systems.. people are scared of the (EE) msgs on the logs ;) | 14:19 |
| apw | so acpid is not used to handle vt switching when KMS is _not_ enabled | 14:20 |
| jcristau | no | 14:24 |
| mvo | tjaalton: virtual (kvm), but still pretty scary | 14:27 |
| * mvo tries to reproduce | 14:28 | |
| tjaalton | mvo: well, as Ng said it's likely gdm restarting | 14:29 |
| jcristau | could be X crashing, could be gdm killing it, or could be the session exiting | 14:29 |
| tjaalton | unless there's something in the logs | 14:29 |
| mvo | tjaalton: yeah, I'm trying to find some traces, unfortunately the previous was done with kvm -snapshot (I wanted to test something different) so I need to reproduce it first to get to the logs | 14:31 |
| === tjaalton_ is now known as tjaalton | ||
| === federico1 is now known as f_bbiab | ||
| === jldugger is now known as pwnguin | ||
| bryce | tjaalton, what do you think of bug 444438? | 18:25 |
| ubottu | Launchpad bug 444438 in fglrx-installer "FGLRX: Some settings files do not named rightly in fglrx driver (8.660) in Karamic Koala" [Undecided,New] https://launchpad.net/bugs/444438 | 18:25 |
| tjaalton | bryce: looks like the package was not installed properly for him | 18:36 |
| tjaalton | ie. interrupted or so | 18:36 |
| bryce | hmm | 18:36 |
| bryce | bug 440233 also seems to be same/similar | 18:37 |
| ubottu | Launchpad bug 440233 in fglrx-installer "fglrx fails at startup because of missing amdpcsdb.default + removal leaves bad settings in Xorg.conf" [High,Triaged] https://launchpad.net/bugs/440233 | 18:37 |
| bryce | fglrx is working fine for me though | 18:37 |
| bryce | and I'd expect more bug reports if there was a real issue here | 18:37 |
| tjaalton | dpkg.log should tell | 18:38 |
| superm1 | maybe something is happening on upgrade for them | 18:39 |
| bryce | the bug report starts with "After messing around with ATI graphics drivers all day..." so god only knows how they got into this situation | 18:40 |
| tjaalton | hehe :) | 18:40 |
| superm1 | well so what's the situation that normally causes files to be installed w/ a dpkg-dist extension? | 18:41 |
| superm1 | there is logic in fglrx's preinst that is supposed to move the old files out of the way in the event they were modified at all so that the new ones can be properly installed | 18:42 |
| superm1 | see xorg-driver-fglrx.preinst | 18:42 |
| tjaalton | I think the package is unconfigured | 18:42 |
| superm1 | if the package was unconfigured, there should have been larger problems though too shouldn't there? | 18:43 |
| tjaalton | sure, but "messing around .. all day" reflects that ;) | 18:43 |
| tjaalton | s/but/and/ | 18:44 |
| bryce | yeah that's what I'm thinking | 18:44 |
| bryce | superm1, https://bugs.edge.launchpad.net/ubuntu/+source/fglrx-installer/+bug/440233/comments/3 | 18:46 |
| ubottu | Launchpad bug 440233 in fglrx-installer "fglrx fails at startup because of missing amdpcsdb.default + removal leaves bad settings in Xorg.conf" [High,Triaged] | 18:46 |
| superm1 | not on karmic, no | 18:46 |
| superm1 | karmic has a prerelease, duh. | 18:47 |
| bryce | ah right | 18:48 |
| superm1 | the original bug in his situation would have been great to see though | 18:48 |
| bryce | ok, I'm just going to close the bug as invalid; I'm reasonably convinced it's just a busted installation. thanks | 18:52 |
| lesshaste | bryce: hi.. are you about? | 19:41 |
| lesshaste | I am looking for a workaround for http://bugs.freedesktop.org/attachment.cgi?id=22595 | 19:44 |
| lesshaste | can I add some mode lines to xorg.conf maybe? | 19:44 |
| === f_bbiab is now known as federico1 | ||
| lesshaste | bryce: I am trying to find a workaround for http://bugs.freedesktop.org/attachment.cgi?id=22595 | 21:01 |
| lesshaste | bryce: I added a Mode line to xorg.conf but that didn't do it | 21:01 |
| bryce | huh? | 21:02 |
| lesshaste | bryce: any ideas how to avoid the null pointer? | 21:02 |
| lesshaste | bryce: I am suffering from the bug that that patch fixes | 21:02 |
| lesshaste | you kindly triaged it for me on ubuntu launchpad | 21:02 |
| lesshaste | I was trying to see if I could workaround the bug easily by editing xorg.conf | 21:02 |
| lesshaste | the patch says "If the screen has no modes defined, can lead to a null pointer dereference in VidModeGetFirstModeline()." | 21:03 |
| lesshaste | so naively I was wondering if I could define some modes to avoid it | 21:03 |
| bryce | lesshaste, why not use the patch? | 21:03 |
| lesshaste | is this wrong/dumb? | 21:03 |
| bryce | probably | 21:03 |
| lesshaste | bryce: I am on intrepid and have never compiled X before | 21:03 |
| lesshaste | and there is no binary for intrepid | 21:03 |
| bryce | intrepid? | 21:03 |
| lesshaste | ubuntu 8.10 | 21:04 |
| bryce | I know what intrepid is | 21:04 |
| lesshaste | you mean why not upgrade? | 21:04 |
| lesshaste | disk space problems at the moment | 21:04 |
| bryce | anyway no, other than applying the patch I do not know of another way to work around it. You could try upgrading. | 21:04 |
| lesshaste | are there simple steps to apply the patch I could follow? | 21:05 |
| bryce | yes, ask on one of the support channels | 21:05 |
| lesshaste | ah.. I sort of thought I was :) | 21:06 |
| lesshaste | sorry to be dumb but the patch comment does imply that the problem is only caused by having no modes defined | 21:38 |
| lesshaste | is this in a different sense to the Mode line in xorg.conf? | 21:38 |
| tormod | lesshaste, can't you just try to make a mode line and you'll see? | 21:43 |
| lesshaste | tormod: I did | 21:43 |
| lesshaste | it made no difference | 21:43 |
| tormod | well there you go | 21:43 |
| lesshaste | this either implies a) that the patch is not relevant to my problem, or b) I don't know how to make a mode line or c) the comment in the patch is not correct | 21:44 |
| tormod | it says "the screen" but that does not have to be the screen you define in xorg.conf | 21:46 |
| lesshaste | http://pastebin.com/f7a98aeb is my xorg.conf | 21:46 |
| lesshaste | tormod: ok.. any more ideas? | 21:46 |
| tormod | I only looked at the patch, would be better if you linked to the bug report ;) | 21:47 |
| tormod | your pastebin failed | 21:48 |
| lesshaste | https://bugs.launchpad.net/bugs/442395 | 21:48 |
| ubottu | Launchpad bug 442395 in xserver-xorg-video-ati "X segfaults with radeon driver - VidModeGetFirstModeline" [Undecided,Confirmed] | 21:48 |
| lesshaste | http://pastebin.ca/1599463 current xorg.conf | 21:49 |
| tormod | are you undocking a laptop? | 21:49 |
| lesshaste | tormod: no | 21:49 |
| lesshaste | just trying to run just about any app :) | 21:50 |
| lesshaste | wine also segfaults X | 21:50 |
| lesshaste | this only started happening in the last month | 21:50 |
| tormod | and the patch does not work? | 21:50 |
| tormod | the last month? on intrepid? | 21:50 |
| lesshaste | tormod: I am still trying to apply it! | 21:50 |
| lesshaste | yes | 21:51 |
| lesshaste | I am running dpkg-buildpackage -b | 21:51 |
| lesshaste | assuming that is the righ thing to do after having applied the patch | 21:51 |
| tormod | yeah that can take a good while | 21:51 |
| tormod | well it sounds more like bug 300310 | 21:52 |
| ubottu | Launchpad bug 300310 in xorg-server "SDL applications crash X server" [High,Fix released] https://launchpad.net/bugs/300310 | 21:52 |
| lesshaste | so I am thinking either it was an intrepid update (there have been quite a few) or it's my xorg.conf | 21:52 |
| lesshaste | but that looks so harmless | 21:52 |
| lesshaste | tormod: ok. The backtrace I posted is almost identical to http://bugs.freedesktop.org/show_bug.cgi?id=19948 however | 21:53 |
| ubottu | Freedesktop bug 19948 in DDX/xorg "segfault in mode selection -- VidModeGetNumOfModes()" [Major,Resolved: duplicate] | 21:53 |
| lesshaste | and I should say that adding Option "DisableVidModeExtension" does stop the segfault | 21:54 |
| lesshaste | it also stops wine working completely | 21:54 |
| tormod | was there an xorg-server-core or -ati update in Intrepid? | 21:54 |
| lesshaste | how do I tell? | 21:54 |
| tormod | you can look through /var/log/dpkg.log | 21:55 |
| tormod | or http://news.gmane.org/gmane.linux.ubuntu.devel.changes.intrepid/ | 21:57 |
| lesshaste | looks like no relevant changes there recently | 21:58 |
| lesshaste | at least not by that name | 21:59 |
| lesshaste | ok build finished | 22:00 |
| lesshaste | hmm.. | 22:00 |
| lesshaste | what do I do know :) | 22:00 |
| lesshaste | I can't seem to see a .deb file | 22:00 |
| lesshaste | s/know/now | 22:02 |
| tormod | it's quite irritating that you keep asking about the same things in many channels | 22:03 |
| bryce | heh | 22:06 |
| lesshaste | bryce: I applied the patch and now I get this X Error of failed request: XF86VidModeExtensionDisabled | 22:07 |
| tormod | lesshaste, and this is not because you have disabled it in xorg.conf as you wrote above? | 22:07 |
| tormod | lesshaste, it is considered impolite and against IRC rules to ask the same question in different channels (and to different people) | 22:08 |
| lesshaste | sorry.. | 22:09 |
| lesshaste | my current xorg.conf | 22:09 |
| lesshaste | http://pastebin.com/f4e838e26 | 22:09 |
| tormod | you are stealing precious time from people who try to help you | 22:09 |
| lesshaste | my current Xorg.0.log http://pastebin.com/f39cf3d9e | 22:09 |
| lesshaste | tormod: understood | 22:10 |
| tormod | lesshaste, I would try without an xorg.conf | 22:11 |
| lesshaste | same message without an xorg.conf | 22:13 |
| lesshaste | just now the resolution is lower :) | 22:13 |
| tormod | where is that message? | 22:13 |
| lesshaste | http://pastebin.ca/1599560 | 22:14 |
| lesshaste | I am totally mystified | 22:16 |
| lesshaste | I applied the patch to xorg-server-1.5.2 | 22:16 |
| lesshaste | and it applied without error | 22:16 |
| lesshaste | tormod: in the meantime my bug has been marked as a duplicate of https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/300310 | 22:18 |
| ubottu | Launchpad bug 300310 in xorg-server "SDL applications crash X server" [High,Fix released] | 22:18 |
| lesshaste | so maybe the patch was just completely the wrong thing to apply? | 22:18 |
| tormod | lesshaste, no I think that's the same bug. identical stacktrace | 22:19 |
| lesshaste | ok | 22:19 |
| tormod | you don't get stacktrace any longer right? the patch worked | 22:19 |
| lesshaste | that's right. it stops the segfaults | 22:20 |
| lesshaste | In a similar way to the Option I mentioned early. | 22:21 |
| lesshaste | but it breaks the apps | 22:21 |
| lesshaste | the key question is why is the mode list returned empty? | 22:21 |
| lesshaste | pScrn = xf86Screens[scrnIndex]; is the line I think that we need to understand | 22:22 |
| lesshaste | xf86.h:extern ScrnInfoPtr *xf86Screens;/* List of pointers to ScrnInfoRecs */ | 22:23 |
| tormod | because there is no screen at that exact time I guess | 22:26 |
| tormod | with xrandr it is possible to turn off all screens | 22:26 |
| bryce | tormod, btw looks like we're not going to get mesa 7.6 in karmic | 22:26 |
| lesshaste | is http://bugs.freedesktop.org/attachment.cgi?id=22595 exactly 157_check_null_modes.patch ? I can't see how to check that | 22:26 |
| jcristau | bryce: what about server 1.6.5? | 22:26 |
| bryce | jcristau, well we had a FFe and everything for mesa, so if that was rejected I'm doubtful anything else is going to get in | 22:27 |
| jcristau | well the server has far less changes, all bugfixes | 22:27 |
| jcristau | s/less/fewer/ | 22:27 |
| tormod | bryce, well that sucks | 22:28 |
| bryce | jcristau, I think pitti would advocate cherrypicking the fixes | 22:29 |
| tormod | so somebody does not trust upstream to make a proper release, and would rather ship an old git master snapshot? | 22:29 |
| bryce | he tends to be rather a stickler about not allowing changes that don't close high priority bug reports in LP | 22:30 |
| bryce | tormod, that's the gist | 22:30 |
| lesshaste | tormod: I have posted to the bug but I don't know if it will help. Any ideas from your end before I give up for the night? | 22:31 |
| tormod | lesshaste, which bug report? | 22:31 |
| lesshaste | tormod: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/442395 | 22:33 |
| ubottu | Launchpad bug 442395 in xserver-xorg-video-ati "X segfaults with radeon driver - VidModeGetFirstModeline (dup-of: 300310)" [Undecided,Confirmed] | 22:33 |
| ubottu | Launchpad bug 300310 in xorg-server "SDL applications crash X server" [High,Fix released] | 22:33 |
| albert23 | lesshaste: Do you have package x11proto-xf86vidmode-dev installed in your build environment? | 22:37 |
| lesshaste | yes | 22:37 |
| lesshaste | can the X log shows (II) Loading extension XFree86-VidModeExtension | 22:38 |
| lesshaste | and the... | 22:38 |
| tormod | lesshaste, is avidemux an SDL app? run ldd /usr/bin/avidemux and look for libsdl something | 22:52 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!