On Thu, 2009-04-16 at 22:22 +0000, Sebastien Bacher wrote:
> valgrind error log
>
> ==13665== Invalid read of size 4
> ==13665== at 0x74E2933: on_checkbutton_share_folder_toggled
> (nautilus-share.c:810)
> ==13665== by 0x46983A3: g_cclosure_marshal_VOID__VOID (gmarshal.c:77)
> ==13665== by 0x468AC7A: g_closure_invoke (gclosure.c:767)
> ==13665== by 0x46A0E56: signal_emit_unlocked_R (gsignal.c:3247)
> ==13665== by 0x46A24B8: g_signal_emit_valist (gsignal.c:2980)
> ==13665== by 0x46A2935: g_signal_emit (gsignal.c:3037)
> ==13665== by 0x42E0459: gtk_toggle_button_toggled (gtktogglebutton.c:365)
> ==13665== by 0x42E04A1: gtk_toggle_button_clicked (gtktogglebutton.c:498)
> ==13665== by 0x46983A3: g_cclosure_marshal_VOID__VOID (gmarshal.c:77)
> ==13665== by 0x46893D8: g_type_class_meta_marshal (gclosure.c:878)
> ==13665== by 0x468AC7A: g_closure_invoke (gclosure.c:767)
> ==13665== by 0x46A06BF: signal_emit_unlocked_R (gsignal.c:3177)
> ==13665== Address 0x932c178 is 16 bytes inside a block of size 72 free'd
> ==13665== at 0x4025DFA: free (vg_replace_malloc.c:323)
> ==13665== by 0x4737125: g_free (gmem.c:190)
> ==13665== by 0x74E2F71: free_property_page_cb (nautilus-share.c:854)
> ==13665== by 0x4719231: g_datalist_clear (gdataset.c:120)
> ==13665== by 0x468F25F: g_object_finalize (gobject.c:747)
> ==13665== by 0x423AF5B: gtk_object_finalize (gtkobject.c:450)
> ==13665== by 0x4331EB8: gtk_widget_finalize (gtkwidget.c:7953)
> ==13665== by 0x468CDF2: g_object_unref (gobject.c:2421)
> ==13665== by 0x423AD0D: gtk_object_destroy (gtkobject.c:406)
> ==13665== by 0x422D377: gtk_notebook_forall (gtknotebook.c:4000)
> ==13665== by 0x4187135: gtk_container_foreach (gtkcontainer.c:1526)
> ==13665== by 0x4187A07: gtk_container_destroy (gtkcontainer.c:1066)
>
> using the current jaunty version
>
Thanks, that helped a lot. I see the error, though can't reproduce it
myself. Basically there's a pointer that's free'd (in
free_property_page_cb) before being accessed in
on_checkbutton_share_folder_toggled.
I see two methods of approaching it:
* Disabling the cancel button (hence the dialog won't be closed
until after the synaptic run completes)
* Allowing the dialog to be cancelled, but making sure there isn't
another attempt to access the free'd data.
Which of the two is more favourable?
--
Regards,
Chow Loong Jin
--
nautilus crashed with SIGSEGV in strlen()
https://bugs.launchpad.net/bugs/258570
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
