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 546494 - basetransform changes cause crash in farsight2 python test
basetransform changes cause crash in farsight2 python test
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gstreamer (core)
git master
Other Linux
: Normal blocker
: NONE
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2008-08-06 00:45 UTC by Olivier Crête
Modified: 2008-10-04 21:50 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Olivier Crête 2008-08-06 00:45:39 UTC 
The newly-committed changes to basetransform cause the farsight2 python gui test to crash in random places.. I believe it must try to access some long-ago freed memory somewhere..

to reproduce you need: the farsight git master (http://git.collabora.co.uk/?p=user/tester/farsight2.git;a=summary) and gst-plugins-farsight 0.12.9 (the latest release) as well as the latest CVS versioin of the various gstreamer packages.

then go into farsight2/tests/gui and run "python fs2-gui.py" .. click on "new server" and wait for crash.. it if does crash, quit and try again (it doesnt happen every time).
Comment 1 Jan Schmidt 2008-08-06 08:08:20 UTC 
Any clue where it's crashing? A stacktrace maybe? :)
Comment 2 Olivier Crête 2008-08-06 14:02:52 UTC 
It crashes in completely random places. I even saw corruption on the stack (near the top).. That said, it seems to crash more often inside ffmpegcsp than in other places, but maybe its not significant.
Comment 3 Tim-Philipp Müller 2008-08-07 16:05:42 UTC 
Maybe valgrind will pick something up?
Comment 4 Olivier Crête 2008-08-20 23:04:32 UTC 
I can't reproduce this anymore with the latest CVS (but it may just be a matter of timing). It may have been the same as bug #548764