Hi Laurent, One more question: On Tue, 6 Dec 2016, Laurent Pinchart wrote: > Hi Guennadi, > > On Tuesday 06 Dec 2016 11:39:22 Guennadi Liakhovetski wrote: > > On Tue, 6 Dec 2016, Laurent Pinchart wrote: > > > On Monday 05 Dec 2016 23:13:53 Guennadi Liakhovetski wrote: > > >> On Tue, 6 Dec 2016, Laurent Pinchart wrote: > > >>>>>> + /* > > >>>>>> + * Register a metadata node. TODO: shall this only be enabled > > >>>>>> for some > > >>>>>> + * cameras? > > >>>>>> + */ > > >>>>>> + if (!(dev->quirks & UVC_QUIRK_BUILTIN_ISIGHT)) > > >>>>>> + uvc_meta_register(stream); > > >>>>>> + > > >>>>> > > >>>>> I think so, only for the cameras that can produce metadata. > > >>>> > > >>>> Every UVC camera produces metadata, but most cameras only have standard > > >>>> fields there. Whether we should stream standard header fields from the > > >>>> metadata node will be discussed later. If we do decide to stream > > >>>> standard header fields, then every USB camera gets metadata nodes. If > > >>>> we decide not to include standard fields, how do we know whether the > > >>>> camera has any private fields in headers without streaming from it? Do > > >>>> you want a quirk for such cameras? > > >>> > > >>> Unless they can be detected in a standard way that's probably the best > > >>> solution. How about a module parameter with a list of VID:PID pairs? The problem with the quirk is, that as vendors produce multiple cameras with different PIDs they will have to push patches for each such camera. Thanks Guennadi -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
