Am Mittwoch, den 13.09.2017, 15:40 -0700 schrieb Rail Shafigulin: > On Wed, Sep 13, 2017 at 2:28 PM, Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote: > > > > There is already a working UVC kernel driver that talks to the correct > > userspace api (which is V4L), in the kernel tree. > > > > So I strongly recommend looking at the code we have today if you have > > questions about it. What exactly is not working for you with what we > > have? > > > > thanks, > > > > greg k-h > > I was not clear about new cameras/camera sensors. If I have a > camera/camera sensor that is not listed in > http://www.ideasonboard.org/uvc/, do I need to write a new v4l2 driver > which in turn will communicate with UVC? Do I modify UVC itself to add > a new camera/camera sensor? Or does the camera/camera sensor need to > comply with some sort of standard in order to be UVC compatible (i.e. > I don't have to write any drivers). Hi, either the camera (to be precise: an interface of it) advertises itself as UVC compliant and implements the UVC standard. If that is the case the existing kernel driver will work and user space can use the V4L API to use the camera. (Theoretically a device can be so stupid that it implements the protocol but does not advertise that. Then you need to add the id to the uvc kernel driver) If the camera does not implent UVC, you will have to write a kernel driver for it, which needs to implement the V4L API. Does this clean up the confusion? Regards Oliver -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
