Bug 622424 – Assertion launching new instance

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 622424 - Assertion launching new instance


Summary:	Assertion launching new instance


Status:	RESOLVED INCOMPLETE

Product:	gtk+
Classification:	Platform
Component:	Widget: Other
Version:	unspecified
Hardware:	Other Linux

Importance:	Normal normal
Target Milestone:	---
Assigned To:	gtk-bugs
QA Contact:	gtk-bugs

URL:
Whiteboard:

Depends on:
Blocks:	622151

Reported:	2010-06-22 18:34 UTC by Bastien Nocera
Modified:	2011-05-16 18:48 UTC

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

Attachments
testapplication.c (1.80 KB, text/plain) 2010-06-22 18:34 UTC, Bastien Nocera	Details

Description Bastien Nocera 2010-06-22 18:34:05 UTC

Created attachment 164332 [details]
testapplication.c

gtk+ master

launch test app attached
launch another instance of test app

In the original instance:
GLib-GObject-CRITICAL **: g_value_get_boxed: assertion `G_VALUE_HOLDS_BOXED (value)' failed

+ Trace 222529

#0 IA__g_logv
at gmessages.c line 555
#1 IA__g_log
at gmessages.c line 569
#2 IA__g_value_get_boxed
at gboxed.c line 548
#3 IA__g_cclosure_marshal_VOID__BOXED
at gmarshal.c line 568
#4 IA__g_closure_invoke
at gclosure.c line 767
#5 signal_emit_unlocked_R
at gsignal.c line 3291
#6 IA__g_signal_emit_valist
at gsignal.c line 2984
#7 IA__g_signal_emit
at gsignal.c line 3041
#8 gtk_application_default_prepare_activation
at gtkapplication.c line 148
#9 IA__g_closure_invoke
at gclosure.c line 767
#10 signal_emit_unlocked_R
at gsignal.c line 3291
#11 IA__g_signal_emit_valist
at gsignal.c line 2984
#12 IA__g_signal_emit
at gsignal.c line 3041
#13 application_dbus_method_call
at gdbusapplication.c line 103
#14 call_in_idle_cb
at gdbusconnection.c line 3778
#15 g_main_dispatch
at gmain.c line 2044
#16 IA__g_main_context_dispatch
at gmain.c line 2597
#17 g_main_context_iterate
at gmain.c line 2675
#18 IA__g_main_loop_run
at gmain.c line 2883
#19 IA__gtk_main
at gtkmain.c line 1185
#20 main
at testapplication.c line 54

Comment 1 Matthias Clasen 2010-06-22 19:23:20 UTC

sounds like a case of wrong marshaller ?
gvariant changed its affiliation from boxed to fundamental...

Comment 2 Christian Persch 2010-06-22 20:30:37 UTC

Looks like a new glib with an old gtk+? I think this is fixed on master, since http://git.gnome.org/browse/gtk+/commit/?id=9d0b941da6af4d5863fdcf731ac702bdf9454df5 . Or perhaps the other way, a new gtk with an old glib?

Comment 3 Bastien Nocera 2010-06-22 20:43:13 UTC

$ rpm -q gtk3 glib2
gtk3-2.90.3-1.fc14.x86_64
glib2-2.25.9-1.fc14.x86_64
glib2-2.25.9-1.fc14.i686

Both are the latest releases.

Comment 4 Christian Persch 2010-06-22 20:53:36 UTC

148  g_signal_emit (G_OBJECT (application), gtk_application_signals[ACTIVATED], 0, arguments);

and the signal is 

  gtk_application_signals[ACTIVATED] =
    g_signal_new (g_intern_static_string ("activated"),
                  G_OBJECT_CLASS_TYPE (klass),
                  G_SIGNAL_RUN_LAST,
                  G_STRUCT_OFFSET (GtkApplicationClass, activated),
                  NULL, NULL,
                  g_cclosure_marshal_VOID__VARIANT,
                  G_TYPE_NONE, 1,
                  G_TYPE_VARIANT);

yet your trace shows a VOID__BOXED marshaler...

Comment 5 Bastien Nocera 2011-05-16 18:48:24 UTC

Let's say this is fixed.