Bug 170083 – nautilus restarts (crashes) when trying to view the properties of a quicktime movie file

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 170083 - nautilus restarts (crashes) when trying to view the properties of a quicktime movie file


Summary:	nautilus restarts (crashes) when trying to view the properties of a quicktime...


Status:	RESOLVED FIXED

Product:	GStreamer
Classification:	Platform
Component:	gst-plugins
Version:	0.8.8
Hardware:	Other Linux

Importance:	Normal normal
Target Milestone:	0.8.9
Assigned To:	Maintainer alias for GStreamer component of Totem
QA Contact:	Maintainer alias for GStreamer component of Totem

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2005-03-12 16:45 UTC by Sebastien Bacher
Modified:	2005-03-20 14:52 UTC

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

Attachments
possible fix (677 bytes, patch) 2005-03-20 11:56 UTC, Ronald Bultje	none	Details \| Review

Description Sebastien Bacher 2005-03-12 16:45:56 UTC

This bug has been opened here: https://bugzilla.ubuntu.com/7475

backtrace of the crash:

(gdb) thread apply all bt

+ Trace 56758

Thread 12 (Thread -1246516304 (LWP 19622))

#0 mute_stream
at gstplaybasebin.c line 769
#1 gst_probe_perform
at gstprobe.c line 112
#2 gst_probe_dispatcher_dispatch
at gstprobe.c line 252
#3 gst_pad_push
at gstpad.c line 3251
#4 gst_qtdemux_loop_header
at qtdemux.c line 818
#5 loop_group_schedule_function
at gstoptimalscheduler.c line 1451
#6 schedule_group
at gstoptimalscheduler.c line 1222
#7 gst_opt_scheduler_schedule_run_queue
at gstoptimalscheduler.c line 1274
#8 schedule_chain
at gstoptimalscheduler.c line 1331
#9 gst_opt_scheduler_iterate
at gstoptimalscheduler.c line 2789
#10 gst_scheduler_iterate
at gstscheduler.c line 744
#11 gst_bin_iterate_func
at gstbin.c line 1246
#12 gst_marshal_BOOLEAN__VOID
at gstmarshal.c line 509
#13 g_cclosure_new_swap
from /usr/lib/libgobject-2.0.so.0
#14 g_closure_invoke
from /usr/lib/libgobject-2.0.so.0
#15 g_signal_emit_by_name
from /usr/lib/libgobject-2.0.so.0
#16 g_signal_emit_valist
from /usr/lib/libgobject-2.0.so.0
#17 g_signal_emit
from /usr/lib/libgobject-2.0.so.0
#18 gst_bin_iterate
at gstbin.c line 1306
#19 gst_thread_main_loop
at gstthread.c line 675
#20 g_static_private_free
from /usr/lib/libglib-2.0.so.0
#21 start_thread
from /lib/tls/i686/cmov/libpthread.so.0
#22 clone
from /lib/tls/i686/cmov/libc.so.6

Comment 1 Sebastien Bacher 2005-03-12 16:55:19 UTC

that happens when gst-ffmpeg is not installed

Comment 2 Ronald Bultje 2005-03-19 14:44:32 UTC

Line 769 contains:
gst_pad_remove_probe ((GstPad *) GST_PAD_REALIZE (info->object), probe);

so, what is info? What is info->object?

Comment 3 Ronald Bultje 2005-03-20 11:56:28 UTC

Created attachment 38958 [details] [review]
possible fix

Sebastien, Federico could reproduce this too... However, I cannot. I've tried a
few different files with different kind of codecs enabled, and it just exits
nicely here. However, from the reports both you and Federico gave me, I think I
have a good indication of what's causing the bug. If I analyse the code
correctly, the attached patch should fix it. What it does is explicitely
synchronize and cleanup the playback thread before destroying the stream-info
structures, which are (later on) used in mute_group to disable the groups
(since we have no codecs for them). With this patch, mute_group should either
not be called alltogether, or should be called before we destroy the
stream-info structures (since we're synchronizing on the thread), which should
in both cases prevent the crash.

Please test and let me know if it helps.

Comment 4 Sebastien Bacher 2005-03-20 14:24:29 UTC

To reproduce you just need to get the nautilus properties for a file not handled
by the plugins (ie: a mov file without ffmpeg installed). BTW the patch fixes
the crash, thanks!

Comment 5 Ronald Bultje 2005-03-20 14:52:55 UTC

OK, re-assign and close then.