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 656771 - invalid read in gst_pad_push_event()
invalid read in gst_pad_push_event()
Status: RESOLVED NOTABUG
Product: GStreamer
Classification: Platform
Component: gstreamer (core)
git master
Other Linux
: Normal blocker
: git master
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2011-08-17 19:42 UTC by Jonathon Jongsma
Modified: 2011-08-18 02:34 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
valgrind log (20.09 KB, application/octet-stream)
2011-08-17 19:42 UTC, Jonathon Jongsma
Details
C test case (2.28 KB, text/plain)
2011-08-17 20:29 UTC, Jonathon Jongsma
Details

Description Jonathon Jongsma 2011-08-17 19:42:14 UTC
Created attachment 194075 [details]
valgrind log

I've observed the following warning on the terminal when trying to seek a stream with gst_element_seek():

GStreamer-CRITICAL **: gst_pad_push_event: assertion `GST_IS_EVENT (event)' failed


valgrind log attached which shows an invalid read in gst_pad_push_event()
Comment 1 Tim-Philipp Müller 2011-08-17 19:54:40 UTC
What's the exact pipeline?

Most likely a refcount bug somewhere.

Are you sure the refcount bug is not in your code or in the bindings (if you're using bindings for gst stuff)?

Can you provide a C test case that triggers this? Or does it also happen with the seek example in -base/tests/examples/seek/, or totem, or gst-launch-0.10 playbin2 uri=file:///...  video-sink='navseek ! autovideosink' ?
Comment 2 Jonathon Jongsma 2011-08-17 20:29:41 UTC
Created attachment 194077 [details]
C test case

No, I'm not entirely sure that the problem isn't due to my code.  That said, I can't see any obvious refcount issues.  I've attached a C test case that displays the problem on my machine.
Comment 3 Tim-Philipp Müller 2011-08-17 21:29:49 UTC
Thanks for that. I can reproduce the problem with git of things.
Comment 4 Tim-Philipp Müller 2011-08-17 21:59:05 UTC
It's probably because you're doing the seek from the streaming thread, which isn't really allowed/expected (same for the set_state(pipeline,...) btw).
Comment 5 Jonathon Jongsma 2011-08-18 02:34:03 UTC
Ah! of course.  Novice gstreamer user error.  Sorry for the hasty bug report, and thanks for your help.