| stgraber | bryce_: Any hope of having bug 327484 fixed by release or should I just add a hack in LTSP to force that device to use VESA instead of geode ? | 00:18 |
|---|---|---|
| ubottu | Launchpad bug 327484 in xserver-xorg-video-geode "X crashing with Geode GX2 on Jaunty" [Undecided,New] https://launchpad.net/bugs/327484 | 00:18 |
| bryce_ | tjaalton: when you get a chance please review the patch I posted on bug #340807 | 00:24 |
| ubottu | Launchpad bug 340807 in xorg "Rename /etc/X11/Xsession.d/60x11-localhost to match convention" [Undecided,New] https://launchpad.net/bugs/340807 | 00:24 |
| bryce_ | stgraber: it needs a "bt full" | 00:25 |
| bryce_ | stgraber: I probably won't have time to do a fix for it for jaunty unless it's something trivially easy like a nullptr ref | 00:25 |
| stgraber | bryce_: bt full attached | 01:12 |
| * stgraber should have put it in a file and attached the file | 01:13 | |
| bryce_ | it's ok | 01:14 |
| bryce_ | hmm | 01:19 |
| bryce_ | replying on bug... | 01:21 |
| stgraber | if you need any more information, I have that thin client right next to me with gdb opened on the X server | 01:24 |
| bryce_ | ok yeah | 01:24 |
| stgraber | only limitation is that it doesn't have an harddrive and only has 90MB of RAM so can't store big files in the ramfs | 01:24 |
| bryce_ | ok, reply on bug | 01:28 |
| stgraber | bryce_: attached the gdb output you requested, not sure it's really helpful most only contains line numbers without any real content (function name, ...) | 02:38 |
| bryce_ | stgraber: yeah, you're just stepping in the signal handler code there | 02:41 |
| bryce_ | stgraber: anyway, I guess forward it upstream; probably not something we can fix at the ubuntu level | 02:41 |
| stgraber | ok, so I should force it to VESA for now ? | 02:42 |
| stgraber | I have 800 of these deployed :) so I need them to at least open a gnome session. | 02:43 |
| tjaalton | bryce_: sure | 07:25 |
| tjaalton | bryce_: hmm, the postrm tries to rm 99xorg-localhost, when it was called 60x11-localhost?-) | 08:07 |
| tjaalton | I'll add a comment on the bug | 08:08 |
| bdmurray | bryce_: bug 341720 isn't really in compiz is it? | 22:37 |
| ubottu | Launchpad bug 341720 in compiz "compiz.real crashed with SIGSEGV in glXCreateContext()" [Medium,New] https://launchpad.net/bugs/341720 | 22:37 |
| bryce_ | bdmurray: here's the error - dlopen: /usr/lib/xorg/modules/extensions//libglx.so: undefined symbol: miInitVisualsProc | 22:40 |
| bdmurray | bryce_: right so it's failing since glx is missing? | 22:41 |
| bryce_ | bdmurray: no; either the user tried to load fglrx (which isn't available yet for Jaunty), or they tried to load radeon but still had fglrx kernel module junk installed | 22:43 |
| bryce_ | so either way it's a user configuration issue, not compiz | 22:43 |
| bryce_ | both problems are already reported, so whichever issue it is, the bug's a dupe | 22:43 |
| bryce_ | bdmurray: from the log it sounds like an xorg.conf was used, but none is attached the the bug; that would be sufficient to determine which issue it was | 22:45 |
| bdmurray | bryce_: the compiz package hook doesn't grab that, do you think it should? | 22:45 |
| bryce_ | yep | 22:46 |
| bryce_ | whatever the case is, you can refer them to https://wiki.ubuntu.com/X/Troubleshooting/FglrxInteferesWithRadeonDriver since that's what they need | 22:46 |
| bryce_ | bdmurray: probably you also want to look in the kernel modules to see if fglrx or nvidia is loaded, and flag those situations, since those bugs will likely be unfixable by us | 22:47 |
| bdmurray | I think apport flags those kernel modules anyway | 22:47 |
| bryce_ | e.g.: | 22:47 |
| bryce_ | try: | 22:47 |
| bryce_ | script = subprocess.Popen(['grep', 'fglrx', '/var/log/kern.log', '/proc/modules'], stdout=subprocess.PIPE) | 22:47 |
| bryce_ | matches = script.communicate()[0] | 22:47 |
| bryce_ | if (matches): | 22:47 |
| bryce_ | report['fglrx-loaded'] = matches | 22:47 |
| bryce_ | except OSError: | 22:47 |
| bryce_ | pass | 22:47 |
| bdmurray | NonfreeKernelModules: fglrx | 22:48 |
| bryce_ | hmm, where's that flagged? | 22:49 |
| bryce_ | I see it in ProcModules.txt and Xorg.0.log | 22:49 |
| bdmurray | right in the description see bug 342501 as an example | 22:49 |
| ubottu | Launchpad bug 342501 in firefox-3.0 "Work offline in ProfileManager does nothing" [Medium,Confirmed] https://launchpad.net/bugs/342501 | 22:49 |
| bdmurray | oh, its not flagged in the compiz bug we are looking at | 22:50 |
| bryce_ | ok right, but for some reason it's not flagged on ... right | 22:50 |
| bdmurray | because no Nonfree ones are in use | 22:50 |
| bryce_ | but "fglrx" is shown in http://launchpadlibrarian.net/23780003/ProcModules.txt | 22:50 |
| bryce_ | fglrx is also in http://launchpadlibrarian.net/23780008/XorgLog.txt | 22:51 |
| bryce_ | so both the kernel module and X driver module are installed and being loaded. Not sure why it's not being flagged | 22:51 |
| bryce_ | you can see though that the fglrx version is wrong... in the Xorg.0.log it complains: | 22:52 |
| bryce_ | [atiddxSetup] X version mismatch - detected X.org 7.1.-1.902, required X.org 7.4.-1.906 | 22:52 |
| bryce_ | I'm interpreting that to mean "Hey, this driver is compiled for X.org 7.1 but we need 7.4" | 22:53 |
| bryce_ | anyway, X600 should be quite finely supported with -ati, no need for fglrx | 22:54 |
| bryce_ | tjaalton: http://people.ubuntu.com/~bryce/totals.svg | 22:56 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
