Hi Hans, On Mon, Jan 22, 2018 at 01:31:16PM +0100, Hans Verkuil wrote: > From: Hans Verkuil <hans.verkuil@xxxxxxxxx> > > There are currently two subdev ops variants to get/set the frame interval: > g/s_parm and g/s_frame_interval. > > This patch series replaces all g/s_parm calls by g/s_frame_interval. > > The first patch adds helper functions that can be used by bridge drivers. > Only em28xx can't use it and it needs custom code (it uses v4l2_device_call() > to try all subdevs instead of calling a specific subdev). > > The next patch converts all non-staging drivers, then come Sakari's > atomisp staging fixes. > > The v4l2-subdev.h patch removes the now obsolete g/s_parm ops and the > final patch clarifies the documentation a bit (the core allows for > _MPLANE to be used as well). > > I would really like to take the next step and introduce two new ioctls > VIDIOC_G/S_FRAME_INTERVAL (just like the SUBDEV variants that already > exist) and convert all bridge drivers to use that and just have helper > functions in the core for VIDIOC_G/S_PARM. > > I hate that ioctl and it always confuses driver developers. It would > also prevent the type of abuse that was present in the atomisp driver. > > But that's for later, let's simplify the subdev drivers first. Apart from my patches, Acked-by: Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> -- Sakari Ailus sakari.ailus@xxxxxxxxxxxxxxx
