Bug 711632 – The desktop-app-info test fails during make check

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 711632 - The desktop-app-info test fails during make check


Summary:	The desktop-app-info test fails during make check


Status:	RESOLVED FIXED

Product:	glib
Classification:	Platform
Component:	gio
Version:	unspecified
Hardware:	Other Linux

Importance:	Normal normal
Target Milestone:	---
Assigned To:	gtkdev
QA Contact:	gtkdev

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2013-11-07 18:52 UTC by Kjartan Maraas
Modified:	2013-11-07 19:04 UTC

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

Attachments
GAppInfoMonitor: don't try to use NULL hash table (1.46 KB, patch) 2013-11-07 18:57 UTC, Allison Karlitskaya (desrt)	committed	Details \| Review

Description Kjartan Maraas 2013-11-07 18:52:57 UTC

Here's the error message  and backtrace:

(/home/kmaraas/src/gnome/glib/gio/tests/.libs/lt-desktop-app-info:26585): GLib-CRITICAL **: g_hash_table_iter_init: assertion 'hash_table != NULL' failed

Program received signal SIGTRAP, Trace/breakpoint trap.
[Switching to Thread 0x7ffff72c4700 (LWP 26589)]
0x00007ffff7d203c1 in g_logv (log_domain=0x7ffff7d81b6e "GLib", 
    log_level=G_LOG_LEVEL_CRITICAL, format=<optimized out>, 
    args=args@entry=0x7ffff72c35d8) at gmessages.c:1019
1019			G_BREAKPOINT ();

+ Trace 232734

Thread 2 (Thread 0x7ffff72c4700 (LWP 26589))

#0 g_logv
at gmessages.c line 1019
#1 g_log
at gmessages.c line 1059
#2 g_return_if_fail_warning
at gmessages.c line 1068
#3 g_app_info_monitor_fire
at gappinfo.c line 1240
#4 ffi_call_unix64
at ../src/x86/unix64.S line 76
#5 ffi_call
at ../src/x86/ffi64.c line 522
#6 g_cclosure_marshal_generic_va
at gclosure.c line 1550
#7 _g_closure_invoke_va
at gclosure.c line 840
#8 g_signal_emit_valist
at gsignal.c line 3238
#9 g_signal_emit
at gsignal.c line 3386
#10 emit_cb
at gfilemonitor.c line 406
#11 g_main_dispatch
at gmain.c line 3066
#12 g_main_context_dispatch
at gmain.c line 3640
#13 g_main_context_iterate
at gmain.c line 3711
#14 g_main_context_iteration
at gmain.c line 3772
#15 glib_worker_main
at gmain.c line 5475
#16 g_thread_proxy
at gthread.c line 798
#17 start_thread
at pthread_create.c line 309
#18 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 111

Comment 1 Allison Karlitskaya (desrt) 2013-11-07 18:57:53 UTC

Created attachment 259219 [details] [review]
GAppInfoMonitor: don't try to use NULL hash table

Don't g_hash_table_iter_init() on a NULL table if we see the app info
monitoring but no monitors have been created yet.

Comment 2 Colin Walters 2013-11-07 19:00:24 UTC

Review of attachment 259219 [details] [review]:

Patch looks obviously right.

Comment 3 Allison Karlitskaya (desrt) 2013-11-07 19:04:22 UTC

Attachment 259219 [details] pushed as bda60c3 - GAppInfoMonitor: don't try to use NULL hash table