GNOME Bugzilla – Bug 356520
secondary screen returns no sizes
Last modified: 2009-02-26 02:03:42 UTC
Version: 2.15.92 What were you doing when the application crashed? opening the application. prgrams running were BonEcho, irssi in a gnome-terminal and gedit Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.15.92 2006-08-22 (Ubuntu) BugBuddy Version: 2.15.92 Memory status: size: 26484736 vsize: 0 resident: 26484736 share: 0 rss: 7409664 rss_rlim: 0 CPU usage: start_time: 1158579017 rtime: 0 utime: 11 stime: 0 cutime:9 cstime: 0 timeout: 2 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/gnome-display-properties' (no debugging symbols found) Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1224558928 (LWP 10832)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 72636
Thread 1 (Thread -1224558928 (LWP 10832))
*** Bug 366386 has been marked as a duplicate of this bug. ***
*** Bug 368616 has been marked as a duplicate of this bug. ***
confirming as per duplciates
*** Bug 374522 has been marked as a duplicate of this bug. ***
*** Bug 375838 has been marked as a duplicate of this bug. ***
*** Bug 379862 has been marked as a duplicate of this bug. ***
*** Bug 388706 has been marked as a duplicate of this bug. ***
*** Bug 390662 has been marked as a duplicate of this bug. ***
*** Bug 391924 has been marked as a duplicate of this bug. ***
Does this still happen? If so, can anybody get us a stack trace with debugging symbols? Please see http://live.gnome.org/GettingTraces for more information on how to do so. Thanks in advance!
*** Bug 415393 has been marked as a duplicate of this bug. ***
*** Bug 416124 has been marked as a duplicate of this bug. ***
*** Bug 420122 has been marked as a duplicate of this bug. ***
(In reply to comment #13) > *** Bug 420122 has been marked as a duplicate of this bug. *** > My problem was caused by weirdness with a video driver, unrelated to this software. Please delete this bug report. Thank you.
Even if it was caused by some weirdness in a video driver we should try not to crash, so this is still relevant. If this weirdness is reproducible for you, getting us a stacktrace with debugging symbols would be much appreciated. In addition, it would be nice if you could tell us with which video driver this weirdness happened (and possibly how), so we can at least check whether the cause is the same for others who are seeing this crash. Thank you.
*** Bug 422644 has been marked as a duplicate of this bug. ***
*** Bug 425896 has been marked as a duplicate of this bug. ***
*** Bug 432459 has been marked as a duplicate of this bug. ***
As a newb to Linux all I can tell you is the following: 1. Dual monitor setup 2. Primary monitor is using onboard Intel graphics / 'i810' driver 3. Secondary monitor is using older PCI nvidia GeForce card / 'nv' driver 4. Xinerama is 'enabled' via xorg.conf - ServerFlags 'Xinerama' 'enabled' ...where are the questions i'm being asked to answer???
We need a stacktrace with debugging symbols. If you follow the link in comment 10, you'll find instructions on how to obtain such a trace for several distributions.
(In reply to comment #19) > As a newb to Linux all I can tell you is the following: > > 1. Dual monitor setup > 2. Primary monitor is using onboard Intel graphics / 'i810' driver > 3. Secondary monitor is using older PCI nvidia GeForce card / 'nv' driver > 4. Xinerama is 'enabled' via xorg.conf - ServerFlags 'Xinerama' 'enabled' > > ...where are the questions i'm being asked to answer??? > > Edit: Works ok with Xinerama enabled... crashes when Xinerama is NOT enabled and > multiple instances of X are running.
(In reply to comment #19) As a newb to Linux all I can tell you is the following: 1. Dual monitor setup 2. Primary monitor is using onboard Intel graphics / 'i810' driver 3. Secondary monitor is using older PCI nvidia GeForce card / 'nv' driver 4. Xinerama is 'enabled' via xorg.conf - ServerFlags 'Xinerama' 'enabled' ...where are the questions i'm being asked to answer??? Edit: Works ok with Xinerama enabled... crashes when Xinerama is NOT enabled and multiple instances of X are running.
npeelman, please see comment #10 in this report. a trace would be really helpful for the developers. thanks.
Here's a backtrace: Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.1 2006-10-02 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 34402304 vsize: 0 resident: 34402304 share: 0 rss: 10182656 rss_rlim: 0 CPU usage: start_time: 1178241855 rtime: 0 utime: 36 stime: 0 cutime:30 cstime: 0 timeout: 6 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/gnome-display-properties' (no debugging symbols found) Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1224959376 (LWP 25021)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 132245
Thread 1 (Thread -1224959376 (LWP 25021))
thank you, that trace is excellent! look like a crash in gtk+. changing product. can you please try to upgrade to Gnome 2.16 or Gnome 2.18 and try to check whether the problem may have been fixed?
I'm not convinced this is gtk territory. I'd like to see a trace including the control-center symbols. npeelman, could supply that as well, please?
New backtrace with gnome-control-center-dbg: Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.1 2006-10-02 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 34398208 vsize: 0 resident: 34398208 share: 0 rss: 10178560 rss_rlim: 0 CPU usage: start_time: 1178333588 rtime: 0 utime: 32 stime: 0 cutime:29 cstime: 0 timeout: 3 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/gnome-display-properties' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1224865168 (LWP 16796)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 132481
Thread 1 (Thread -1224865168 (LWP 16796))
Thanks a lot. Definitely not a GTK problem. npeelman (or somebody else who can reproduce), are you available via IRC/IM? I'd like to step you through a gdb session to fix this. Apparently we're getting fed some unexpected data. Unfortunately, I don't have a dual-monitor setup to test with myself, so I need someone to poke at stuff for me. You can often find me on the #control-center channel on irc.gnome.org (fizz). Feel free to mail me to exchange other contact details if necessary.
FULL backtrace at request of Jens Granseuer: (gdb) bt full
+ Trace 132568
Ok, so for posterity: Thanks to npeelman's relentless efforts we have been able to identify the source of the issue. It appears XRRGetSizes can return 0 (zero) resolutions for an existing screen. That in turn leaves the capplet in an inconsistent state. Now, I'm not sure whether the fact that 0 is returned is actually a driver problem or something else. I'm trying to get an answer to that from the X guys. Hopefully, that will also give us an idea how to handle this situation properly.
Norman, could you please tell us the versions of xorg-server and libXrandr you are using?
According to Synaptic: Installed- xserver-xorg version 1:7.1.1ubuntu6.2 libxrandr2 version 2:1.1.1-0ubuntu1
I just committed a change to fix the crashes. I'm still unsure how we're supposed to properly handle this case, though.
This is obsolete with the new RANDR code.