Em Thu, 30 Mar 2017 23:26:26 +0300 Sakari Ailus <sakari.ailus@xxxxxx> escreveu: > Hi Helen and others, > > On Thu, Mar 30, 2017 at 01:02:18PM -0300, Helen Koike wrote: > > add documentation for V4L2_INPUT_TYPE_DEFAULT > > > > Signed-off-by: Helen Koike <helen.koike@xxxxxxxxxxxxx> > > --- > > Documentation/media/uapi/v4l/vidioc-enuminput.rst | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/Documentation/media/uapi/v4l/vidioc-enuminput.rst b/Documentation/media/uapi/v4l/vidioc-enuminput.rst > > index 17aaaf9..0237e10 100644 > > --- a/Documentation/media/uapi/v4l/vidioc-enuminput.rst > > +++ b/Documentation/media/uapi/v4l/vidioc-enuminput.rst > > @@ -112,6 +112,9 @@ at index zero, incrementing by one until the driver returns ``EINVAL``. > > :stub-columns: 0 > > :widths: 3 1 4 > > > > + * - ``V4L2_INPUT_TYPE_DEFAULT`` > > + - 0 > > + - This is the default value returned when no input is supported. Input *IS* supported. The device has one input. So, the description is wrong ;) > > * - ``V4L2_INPUT_TYPE_TUNER`` > > - 1 > > - This input uses a tuner (RF demodulator). > > What would you think of calling this input as "unknown" instead of > "default"? That's what an input which isn't really specified actually is. Yeah, default seems a bad name to me, too. Actually, I think that the best here would be to not create a new type. just use V4L2_INPUT_TYPE_CAMERA. That's actually the default for the embedded drivers. If a driver is providing input for something else, then it should implement vidioc_enuminput method. Thanks, Mauro
