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 362566 - [dv1394src] Random segfault and kernel oopses
[dv1394src] Random segfault and kernel oopses
Status: RESOLVED NOTGNOME
Product: GStreamer
Classification: Platform
Component: gst-plugins-good
0.10.4
Other All
: Normal critical
: 0.10.6
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2006-10-16 10:56 UTC by Jerome Alet
Modified: 2007-03-02 09:46 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Jerome Alet 2006-10-16 10:56:51 UTC 
Steps to reproduce:
1. Reboot
2. Run the pipeline at http://pastebin.ca/204885
3. Run this one "gst-launch dv1394src use-avc=false ! fakesink"


Stack trace:

+ Trace 76583

Thread 1 (process 2527)

  • #0 memcpy
    from /lib/tls/i686/cmov/libc.so.6
  • #1 iec61883_dv_get_buffers
    from /usr/lib/libiec61883.so.0
  • #2 iec61883_dv_fb_start
    from /usr/lib/libiec61883.so.0
  • #3 _raw1394_iso_iterate
    from /usr/lib/libraw1394.so.8
  • #4 raw1394_loop_iterate
    from /usr/lib/libraw1394.so.8
  • #5 gst_dv1394src_get_type
    from /usr/lib/gstreamer-0.10/libgst1394.so
  • #6 gst_push_src_get_type
    from /usr/lib/libgstbase-0.10.so.0
  • #7 gst_base_src_get_type
    from /usr/lib/libgstbase-0.10.so.0
  • #8 gst_base_src_get_type
    from /usr/lib/libgstbase-0.10.so.0
  • #9 gst_task_set_lock
    from /usr/lib/libgstreamer-0.10.so.0
  • #10 g_thread_pool_push
    from /usr/lib/libglib-2.0.so.0
  • #11 g_thread_create_full
    from /usr/lib/libglib-2.0.so.0
  • #12 start_thread
    from /lib/tls/i686/cmov/libpthread.so.0
  • #13 clone
    from /lib/tls/i686/cmov/libc.so.6 

Other information:
The segfault or kernel oopses don't occur always, it is random. Most of the times, the original pipeline works just fine. However I've noticed that the segfaults often occurs the first or second times I run the original pipeline, then launching the simpler one reproduces the segfault as well. Kernel oopses maybe are unrelated.
Comment 1 Snark 2007-02-21 19:31:05 UTC 
I didn't check your example, but I know the firewire subsystem has always been pretty unstable for me -- even in kino.

Does kino work better than the gstreamer pipeline on your box in the same conditions ?
Comment 2 Jerome Alet 2007-02-21 19:51:32 UTC 
I think this problem was fixed in newer releases since I've not experienced it for a long time, and our big pipeline have been in production (slightly altered) for two months and a half now.

However I agree with you firewire support has always been problematic, and I must say especially in Kino : this software very often segfaults when capturing from firewire, to the point that I don't use it anymore (maybe this was fixed).
Comment 3 Wim Taymans 2007-03-02 09:46:47 UTC 
ok, closing.