After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 340830 - Random crashes with GTK 2.8.12 and above
Random crashes with GTK 2.8.12 and above
Status: RESOLVED DUPLICATE of bug 341327
Product: gtk+
Classification: Platform
Component: Widget: Other
unspecified
Other All
: Normal critical
: ---
Assigned To: gtk-bugs
gtk-bugs
Depends on:
Blocks:
 
 
Reported: 2006-05-06 13:44 UTC by Enrico Tröger
Modified: 2006-06-04 17:22 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
valgrind output while crashing (60.98 KB, text/plain)
2006-05-31 14:21 UTC, Enrico Tröger
Details

Description Enrico Tröger 2006-05-06 13:44:45 UTC
Steps to reproduce:
Sometimes it crashes at startup, sometimes you can start Geany and work for a
few minutes and then it crashes. I don't know any special actions to trigger a
crash.

Stack trace:
This one I get on FreeBSD with GTK 2.8.12:

Program received signal SIGSEGV, Segmentation fault.
0x286222e3 in gdk_x11_register_standard_event_type ()
from /usr/X11R6/lib/libgdk-x11-2.0.so.0
(gdb) bt
  • #0 gdk_x11_register_standard_event_type
  • #0 gdk_x11_register_standard_event_type
  • #5 g_main_loop_run
    from /usr/lib/libglib-2.0.so.0
  • #0 g_type_check_instance_is_a
    from /usr/lib/libgobject-2.0.so.0
  • #1 gdk_x11_register_standard_event_type
    from /usr/lib/libgdk-x11-2.0.so.0
  • #2 _gdk_events_queue
    from /usr/lib/libgdk-x11-2.0.so.0
  • #3 _gdk_events_queue
    from /usr/lib/libgdk-x11-2.0.so.0
  • #4 g_main_context_dispatch
  • #0 gdk_event_translate
    at gdkevents-x11.c line 894
  • #1 _gdk_events_queue
    at #gdkevents-x11.c line 2225
  • #3 IA__g_main_context_dispatch
    at #gmain.c line 1916
  • #5 IA__g_main_loop_run
    at gmain.c line 2751
  • #6 IA__gtk_main
    at gtkmain.c line 1001
  • #7 main
    at main.c line 577


Other information:
The application where the bug occurs is Geany (http://geany.uvena.de). I'm the
developer of this application and some of my users reported the crashes. With
GTK version before 2.8.12 everything is fine and it is stable. Using GTK
versions after 2.8.10 or .11 Geany becomes very unstable and crashes nearly always.
I'm not completely sure, whether this is a GTK bug or whether it is a bug in my
application, but it must be GTK related, since earlier GTK versions work.
(if it is not a GTK bug, sorry for this useless report)
Comment 1 Tim Janik 2006-05-11 22:29:26 UTC
to repeat your last statement:
- GTK < 2.8.12 is stable
- GTK >= 2.8.10 crashes
can you please clarify the overlap for versions .10 and .11?
and, what glib version are you using here?

this issue may be related to bug 341327.
Comment 2 Enrico Tröger 2006-05-12 00:31:33 UTC
(In reply to comment #1)
> to repeat your last statement:
> - GTK < 2.8.12 is stable
> - GTK >= 2.8.10 crashes
> can you please clarify the overlap for versions .10 and .11?
No, I can't. I know, with GTK 2.8.10 and GLib 2.10.2(yes, GLib 2.10) it runs stable and as soon as I upgrade my GTK version to something higher like 2.8.12, it crashes. I can't test GTK 2.8.11 because I have no packages for it, sorry.

> and, what glib version are you using here?
On the Debian testing system:
GTK 2.8.16 and GLib 2.10.2
On the FreeBSD 6 system:
GTK 2.8.17 with GLib 2.8.6

And running Geany with G_SLICE=always-malloc doesn't help at all.

I just downgrade the GLib version to 2.8.6, and it crashes, too.
So, I think the g_slice is not the reason.
Comment 3 Enrico Tröger 2006-05-17 16:11:24 UTC
Any news about the topic?

It also happens with GTK 2.8.17 and GLib 2.10.2.
Comment 4 zenwalk-oskar 2006-05-28 19:41:53 UTC
This is a backtrace from the latest gtk+2.8.18 and glib-2.10.3 (Zenwalk)

Starting program: /usr/bin/geany
[Thread debugging using libthread_db enabled]
[New Thread -1217714496 (LWP 6572)]

Program received signal SIGSEGV, Segmentation fault.

Thread NaN (LWP 6572)

  • #0 gdk_x11_register_standard_event_type
    from /usr/lib/libgdk-x11-2.0.so.0
  • #1 gdk_event_get_graphics_expose
    from /usr/lib/libgdk-x11-2.0.so.0
  • #2 gdk_event_get_graphics_expose
    from /usr/lib/libgdk-x11-2.0.so.0
  • #3 g_main_context_dispatch
    from /usr/lib/libglib-2.0.so.0
  • #4 g_main_context_acquire
    from /usr/lib/libglib-2.0.so.0
  • #5 g_main_loop_run
    from /usr/lib/libglib-2.0.so.0
  • #6 gtk_main
    from /usr/lib/libgtk-x11-2.0.so.0
  • #7 main
    at main.c line 569

Comment 5 Tim Janik 2006-05-31 00:00:38 UTC
since gdk_x11_register_standard_event_type() is called in your stack trace for graphics expose, it might be something in your program, since such a function should usually run stable (and does on a lot of systems).

can you possibly strip down your test case to a short program we could use for reproduction?
Comment 6 Enrico Tröger 2006-05-31 14:19:16 UTC
Sorry, I can't. I don't know what to strip down because I don't know where the bug could be. And it is not really reproducable but occurs often enough to see it.
My experience is, opening the preferences dialog of Geany let it crash several times or sometimes just starting it, leads to a crash.

I'll attach the output of valgrind, I just opened some files and typed some text.
Hope this helps.
Comment 7 Enrico Tröger 2006-05-31 14:21:38 UTC
Created attachment 66533 [details]
valgrind output while crashing

This is the output of valgrind starting Geany, opening a few files and typing some text. Then it crashed.
(the output is not complete, the first few lines are missing but they were similar to the first lines in the file)
Comment 8 Enrico Tröger 2006-06-04 16:56:13 UTC
It seems to be solved. The bugfix mentioned in http://mailman.lyra.org/pipermail/scintilla-interest/2006-May/007814.html stopped the crashes.
Geany includes the Scintilla editing component and we changed the Scintilla code included in Geany and now it seems to work.

The problem was, that Scintilla mixed X core fonts with Pango fonts.

Thanks for your help.
Comment 9 Tim Janik 2006-06-04 17:22:21 UTC
closing bug as duplicate of #341327.

*** This bug has been marked as a duplicate of 341327 ***