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 101037 - Ogg related Crash with gstreamer 0.5.0
Ogg related Crash with gstreamer 0.5.0
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: dont know
0.5.0
Other Linux
: Normal major
: 0.3.3
Assigned To: RhythmBox Maintainers
RhythmBox Maintainers
: 101655 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2002-12-12 16:39 UTC by Frederic Crozat
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Frederic Crozat 2002-12-12 16:39:12 UTC 
These backtrace are from cooker packages ie RB 0.4.1, Monkey-media 0.6.1
and gstreamer 0.5.0

on an ogg file:

0x40b90e7c in memcpy () from /lib/i686/libc.so.6

+ Trace 31616

  • #1 gst_bytestream_assemble
    at bytestream.c line 351
  • #2 gst_bytestream_peek_bytes
    at bytestream.c line 317
  • #3 gst_vorbisfile_read
    at vorbisfile.c line 247
  • #4 ov_comment
    from /usr/lib/libvorbisfile.so.3
  • #5 ov_comment
    from /usr/lib/libvorbisfile.so.3
  • #6 ov_read
    from /usr/lib/libvorbisfile.so.3
  • #7 gst_vorbisfile_loop
    at vorbisfile.c line 574
  • #8 gst_basic_scheduler_loopfunc_wrapper
    at gstbasicscheduler.c line 277
  • #9 cothread_stub
    at cothreads.c line 493
  • #0 memcpy
    from /lib/i686/libc.so.6
  • #1 gst_mad_chain
    at gstmad.c line 1021
  • #2 gst_basic_scheduler_chain_wrapper
    at gstbasicscheduler.c line 335
  • #3 cothread_stub
    at cothreads.c line 493 






Comment 1


Olivier Martin





          2002-12-13 12:44:45 UTC
        



Both backtraces look like a GStreamer problem.






Comment 2


David Kennedy





          2002-12-20 16:14:43 UTC
        



*** Bug 101655 has been marked as a duplicate of this bug. ***






Comment 3


Thomas Vander Stichele





          2003-01-05 16:57:21 UTC
        



I think we confirmed this was an optimization issue, no ?
What where the opt flags used in compiling ?






Comment 4


Frederic Crozat





          2003-01-06 08:54:14 UTC
        



Default optimizations for Mandrake are now :  -O2 -fomit-frame-pointer
-pipe -march=i586 -mcpu=pentiumpro

But this stacktrace was produced with a -O1 -g  build :((

All of this with gcc 3.2.1






Comment 5


Christian Fredrik Kalager Schaller





          2003-02-24 12:28:49 UTC
        



this seems scheduler related, CVS now use opt as default and shortly
upcoming 0.6.1 will ship with opt as default.






Comment 6


Christian Fredrik Kalager Schaller





          2003-04-07 17:24:35 UTC
        



Haven't had any updated on this bug since switch to opt. Closing.