Re: [PATCH 10/10] venus: dec: make decoder compliant with stateful codec API

[Tomasz Figa]

On Wed, Jan 30, 2019 at 12:18 PM Nicolas Dufresne <nicolas@xxxxxxxxxxxx> wrote:
>
> Le lundi 28 janvier 2019 Ã 16:38 +0900, Tomasz Figa a Ãcrit :
> > > > Nope, that's not what is expected to happen here. Especially since
> > > > you're potentially in non-blocking IO mode. Regardless of that, the
> > >
> > > OK, how to handle that when userspace (for example gstreamer) hasn't
> > > support for v4l2 events? The s5p-mfc decoder is doing the same sleep in
> > > g_fmt.
> >
> > I don't think that sleep in s5p-mfc was needed for gstreamer and
> > AFAICT other drivers don't have it. Doesn't gstreamer just set the
> > coded format on OUTPUT queue on its own? That should propagate the
> > format to the CAPTURE queue, without the need to parse the stream.
>
> Yes, unfortunately, GStreamer still rely on G_FMT waiting a minimal
> amount of time of the headers to be processed. This was how things was
> created back in 2011, I could not program GStreamer for the future. If
> we stop doing this, we do break GStreamer as a valid userspace
> application.

Does it? Didn't you say earlier that you end up setting the OUTPUT
format with the stream resolution as parsed on your own? If so, that
would actually expose a matching framebuffer format on the CAPTURE
queue, so there is no need to wait for the real parsing to happen.

>
> This is not what I want long term, but I haven't got time to add event
> support, and there is a certain amount of time (years) when this is
> implemented before all the old code goes away.
>
> Nicolas
>