Bug 3766 - v4l Xv overlay output stops working after a few hours.
Summary: v4l Xv overlay output stops working after a few hours.
Status: RESOLVED NOTOURBUG
Alias: None
Product: xorg
Classification: Unclassified
Component: * Other (show other bugs)
Version: 6.8.2
Hardware: x86 (IA32) Linux (All)
: high normal
Assignee: Xorg Project Team
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-07-13 12:17 UTC by Aaron VanDevender
Modified: 2005-07-14 14:18 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments

Description Aaron VanDevender 2005-07-13 12:17:16 UTC
I'm running an app (cupid) which uses a v4l Xv to display video from a capture
card. I have the v4l module loaded. If I start the program soon after X is
started it works fine. But if I come back a while (several hours) later and run
cupid, it only shows a black rectangle where the video is supposed to be.
Restarting X allows cupid to function correctly again.

I'm running FC4 with the nvidia (proprietary) driver.
Comment 1 Aaron VanDevender 2005-07-14 08:28:04 UTC
I compared values of xvinfo from while Xv v4l is working and after it chokes,
and the thing that stuck out at me is that before it stops wowrking XV_FREQ in
Adapter #0 "video4linux" is set to 0. After it stops working its set to 68718
which is outside the allowed range of 0-16000. If XvSetPortAttribute is called
with a value above 16000 it should return BadValue, but for some reason, it is
instead actually setting the erroneous value. I don't know if that is what is
causing my problem, but that certainly isn't helping.
Comment 2 Aaron VanDevender 2005-07-16 00:18:10 UTC
This effect (video not displaying) is due to some applications setting
XV_AUTOPAINT_COLORKEY=0. For example, xine does this, and it has a bug whereby
under some circumstances it dose not set the Xv attribute back to 1 when it
quits. You can set it from the commandline using

xvattr -a XV_AUTOPAINT_COLORKEY -v 1

I still think Xv should refuse to set XV_FREQ to a value outside of its range,
but it appears that has nothing to do with this display problem, so I'm closing
this bug none the less, and I'll go and file a bug with xine, the real perpetrator.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.