Bug 648949 – crash in gvdb_table_lookup at ../gvdb/gvdb-reader.c

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 648949 - crash in gvdb_table_lookup at ../gvdb/gvdb-reader.c


Summary:	crash in gvdb_table_lookup at ../gvdb/gvdb-reader.c


Status:	RESOLVED FIXED

Product:	dconf
Classification:	Core
Component:	gsettings backend
Version:	git master
Hardware:	Other Linux

Importance:	Normal critical
Target Milestone:	---
Assigned To:	dconf-maint
QA Contact:	dconf-maint

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2011-04-29 15:29 UTC by Matthias Clasen
Modified:	2011-05-05 13:44 UTC

See Also:
GNOME target:	---
GNOME version:	---

Description Matthias Clasen 2011-04-29 15:29:03 UTC

https://bugzilla.redhat.com/show_bug.cgi?id=689984

Here we see dconf calling munmap in one thread, and another thread gets a segfault in gvdb_table_lookup

Comment 1 Matthias Clasen 2011-04-29 15:29:24 UTC

Direct link to the stacktrace: https://bugzilla.redhat.com/attachment.cgi?id=486921

Comment 2 Fabio Durán Verdugo 2011-04-29 22:15:37 UTC

Paste here the stacktrace.

warning: core file may not match specified executable file.
[New LWP 3344]
[New LWP 3346]
[New LWP 3345]
[Thread debugging using libthread_db enabled]
Core was generated by `gnome-control-center keyboard'.
Program terminated with signal 11, Segmentation fault.

+ Trace 226947

Thread 1 (Thread 0x7fd27a80d940 (LWP 3344))

#0 gvdb_table_lookup
at ../gvdb/gvdb-reader.c line 265
#1 gvdb_table_get_value
at ../gvdb/gvdb-reader.c line 441
#2 dconf_engine_read
at ../engine/dconf-engine.c line 415
#3 dconf_settings_backend_read
at dconfsettingsbackend.c line 408
#4 dconf_settings_backend_read
at dconfsettingsbackend.c line 394
#5 g_settings_backend_read
at gsettingsbackend.c line 730
#6 g_settings_read_from_backend
at gsettings.c line 1084
#7 g_settings_get_value
at gsettings.c line 1265
#8 g_settings_get_string
at gsettings.c line 1734
#9 gkbd_keyboard_config_load_params
at gkbd-keyboard-config.c line 330
#10 gkbd_keyboard_config_load
at gkbd-keyboard-config.c line 441
#11 ??
from /usr/lib64/control-center-1/panels/libregion.so
#12 ??
from /usr/lib64/control-center-1/panels/libregion.so
#13 g_closure_invoke
at gclosure.c line 767
#14 signal_emit_unlocked_R
at gsignal.c line 3252
#15 g_signal_emit_valist
at gsignal.c line 2983
#16 g_signal_emit
at gsignal.c line 3040
#17 g_settings_real_change_event
at gsettings.c line 280
#18 _gio_marshal_BOOLEAN__POINTER_INT
at gio-marshal.c line 357
#19 g_closure_invoke
at gclosure.c line 767
#20 signal_emit_unlocked_R
at gsignal.c line 3290
#21 g_signal_emit_valist
at gsignal.c line 2993
#22 g_signal_emit
at gsignal.c line 3040
#23 settings_backend_path_changed
at gsettings.c line 341
#24 g_settings_backend_invoke_closure
at gsettingsbackend.c line 265
#25 g_main_dispatch
at gmain.c line 2440
#26 g_main_context_dispatch
at gmain.c line 3013
#27 g_main_context_iterate
at gmain.c line 3091
#28 g_main_loop_run
at gmain.c line 3299
#29 gtk_main
at gtkmain.c line 1358
#30 g_application_run
at gapplication.c line 1322
#31 main

Comment 3 Allison Karlitskaya (desrt) 2011-05-05 13:44:59 UTC

commit da7229b0b2c3a4aec392a042fd231bea7aaecfc0
Author: Ryan Lortie <desrt@desrt.ca>
Date:   Thu May 5 15:38:21 2011 +0200

    Change locking in the engine
    
    Switch from code locking to data locking and hold the data lock quite a
    lot more often.
    
    We could try to be more clever about finding ways to hold locks less
    often but I think it's not really worth it.
    
    Closes #648949