> > I don't like the driver doing to much automagically. > > But you agree that audio is not really useful for read() > capturing? It isn't. I think we have to extend the v4l2 API a bit to handle this. /me can think of two different ways to archive this: * Add some "i don't care about audio" flag to the device context (and a ioctl to set/clear this). If unset, the driver will switch audio input together with video input, otherwise not. * Add a audio input to the device context and reqire applications to set the audio input explicitly the same way they to with the video input (needs a new ioctl too). Have some deadbeef magic for the audio input for the "i don't care" case. > I agree with the things, Bill already sent in another mail. Fine, I've updated the text according to his comments (changes not online yet). Gerd -- Protecting the children is a good way to get a lot of adults who cant stand up for themselves. -- seen in some sig on /.
