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 310894 - videoscale causes segfaults in YUY2 colorspace
videoscale causes segfaults in YUY2 colorspace
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gst-plugins-base
git master
Other Linux
: Normal major
: 0.8.x
Assigned To: Andy Wingo
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2005-07-19 17:06 UTC by Andy Wingo
Modified: 2005-07-20 17:34 UTC
See Also:
GNOME target: ---
GNOME version: ---

Attachments
Valgrinding current CVS, 0.8 branch (26.27 KB, text/plain)
2005-07-20 09:21 UTC, Andy Wingo 		Details

Description Andy Wingo 2005-07-19 17:06:44 UTC 
gst-launch-0.8 videotestsrc ! videoscale !
video/x-raw-yuv,width=240,height=192,framerate=10.0,format=\(fourcc\)YUY2 ! fakesink

I get a segfault during the first iteration. Valgrind also reports fishy things
going on. This problem is also present in 0.9. It is not present in other
colorspaces. Videoscale is my current suspect, although I'm not sure.
Comment 1 Andy Wingo 2005-07-19 17:07:50 UTC 
Sorry to bother you again Dave, but I'd appreciate it if you had a look at this.
Thanks bunches. I'll also be looking at it as well.
Comment 2 David Schleef 2005-07-19 17:41:02 UTC 
Could you be a dear and attach a valgrind log?  Valgrind appears to be borkified
on my machine.
Comment 3 Michael Smith 2005-07-20 08:09:42 UTC 
Wingo asked me to check this one.

This pipeline works for me using version 0.8.8.

If I replace fakesink with xvimagesink, it also works.
Comment 4 Andy Wingo 2005-07-20 09:20:48 UTC 
And mine is 0.8.10/cvs (both). Attaching a valgrind log, which shows some
problems but then segfaults valgrind.
Comment 5 Andy Wingo 2005-07-20 09:21:38 UTC 
Created attachment 49444 [details]
Valgrinding current CVS, 0.8 branch
Comment 6 Andy Wingo 2005-07-20 16:27:54 UTC 
I see the problem only with nearest-neighbor...
Comment 7 Andy Wingo 2005-07-20 17:34:45 UTC 
Fixed in cvs, was a typo in vs_image.c