[Bug 728438] v4l2: Implement a v4l2 video encoder

GStreamer (GNOME Bugzilla) bugzilla at gnome.org
Tue Sep 1 00:56:33 PDT 2015


https://bugzilla.gnome.org/show_bug.cgi?id=728438

--- Comment #52 from Jean-Michel Hautbois (jmleo) <jean-michel.hautbois at veo-labs.com> ---
(In reply to ayaka from comment #51)
> (In reply to Jean-Michel Hautbois (jmleo) from comment #49)
> > ayaka: we have a propose_allocation in our gitlab repo
> > (https://gitlab.com/veo-labs/gst-plugins-good/commit/
> > 52d1aa7d9ba751866c6e5a7b42d1dc0131044071).
> > Are you planning on working on this again ?
> > If so, then, right now there is a bug with the coda encoder (but probably
> > others) which makes it stop after 32 start/stop.
> > 
> > The real issue is that the encoder instance is not closed, because a buffer
> > is leaked.
> > I still have not found the correction, but if you have some time to take a
> > look, it could help.
> 
> I am very sorry about that.
> I didn't meet that problem in my exynos 4412(I have long time not update my
> git repo), do you meet the same problem when you use the same version as
> mine?

I don't know, I have a modified version for videoaggregator as well. The issue
is reproducible only if you switch the pipeline from NULL to PLAYING, then
NULL, PLAYING, etc.
You can't reproduce it with gst-launch as it will close all open files.

> And which hardware do you use?

i.MX6 with coda driver. I don't think it is a driver issue.

> Could you tell me more detail in IRC, I will be there for a few days.

Of course, please ping me.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.


More information about the gstreamer-bugs mailing list