Em Thu, 9 Mar 2017 13:54:45 +0200
Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> escreveu:
> The caller must always call the s_power() op symmetrically powering the
> device on and off. This is the practice albeit it was not documented. A
> lot of sub-device drivers rely on it, so document it accordingly.
Actually, there are several non-embedded drivers that don't call s_power()
symmetrically, as they use it just put a sub-device in standby mode.
I remember some widely used tuners that has this behavior. Such subdevs
automatically awaken when a command is issued to them (for example,
requesting the tuner to tune into a channel).
If you're willing to change the kABI, you need first to patch such
drivers.
Regards,
Mauro
>
> Signed-off-by: Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx>
> ---
> include/media/v4l2-subdev.h | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/include/media/v4l2-subdev.h b/include/media/v4l2-subdev.h
> index 0ab1c5d..b4e521d 100644
> --- a/include/media/v4l2-subdev.h
> +++ b/include/media/v4l2-subdev.h
> @@ -172,8 +172,10 @@ struct v4l2_subdev_io_pin_config {
> *
> * @s_register: callback for %VIDIOC_G_REGISTER ioctl handler code.
> *
> - * @s_power: puts subdevice in power saving mode (on == 0) or normal operation
> - * mode (on == 1).
> + * @s_power: Puts subdevice in power saving mode (on == 0) or normal operation
> + * mode (on == 1). The caller is responsible for calling the op
> + * symmetrically, i.e. calling s_power(dev, 1) once requires later calling
> + * s_power(dev, 0) once.
> *
> * @interrupt_service_routine: Called by the bridge chip's interrupt service
> * handler, when an interrupt status has be raised due to this subdev,
Thanks,
Mauro
