Message ID | 20200904215141.20862-3-luca@lucaceresoli.net (mailing list archive) |
---|---|
State | Superseded |
Headers | show |
Series | [1/3] media: docs: v4l2-subdev: fix typo | expand |
Hi Luca, On Fri, Sep 04, 2020 at 11:51:41PM +0200, Luca Ceresoli wrote: > This paragraph provides generic information to explain what v4l2_subdev is > useful for. Placing it in the middle of paragraphs describing the details > of subdev registration does not make much sense. Move it near the beginning > of the section when the v4l2_subdev idea has just been introduced and > before going into its details. > > Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net> > --- > Documentation/driver-api/media/v4l2-subdev.rst | 12 ++++++------ > 1 file changed, 6 insertions(+), 6 deletions(-) > > diff --git a/Documentation/driver-api/media/v4l2-subdev.rst b/Documentation/driver-api/media/v4l2-subdev.rst > index fb66163deb38..1c1e3f9da142 100644 > --- a/Documentation/driver-api/media/v4l2-subdev.rst > +++ b/Documentation/driver-api/media/v4l2-subdev.rst > @@ -12,6 +12,12 @@ Usually these are I2C devices, but not necessarily. In order to provide the > driver with a consistent interface to these sub-devices the > :c:type:`v4l2_subdev` struct (v4l2-subdev.h) was created. > > +The advantage of using :c:type:`v4l2_subdev` is that it is a generic struct and > +does not contain any knowledge about the underlying hardware. So a driver might > +contain several subdevs that use an I2C bus, but also a subdev that is > +controlled through GPIO pins. This distinction is only relevant when setting > +up the device, but once the subdev is registered it is completely transparent. > + > Each sub-device driver must have a :c:type:`v4l2_subdev` struct. This struct > can be stand-alone for simple sub-devices or it might be embedded in a larger > struct if more state information needs to be stored. Usually there is a > @@ -235,12 +241,6 @@ it can call ``v4l2_subdev_notify(sd, notification, arg)``. This macro checks > whether there is a ``notify()`` callback defined and returns ``-ENODEV`` if not. > Otherwise the result of the ``notify()`` call is returned. > > -The advantage of using :c:type:`v4l2_subdev` is that it is a generic struct and > -does not contain any knowledge about the underlying hardware. So a driver might > -contain several subdevs that use an I2C bus, but also a subdev that is > -controlled through GPIO pins. This distinction is only relevant when setting > -up the device, but once the subdev is registered it is completely transparent. > - Have you considered moving the whole part that describes how to call operations, which comes after the synchronous registration case to a dedicated sub-section ? The above paragraph makes sense in the context of describing why v4l2_subdev is advantageous as it abstract the underlying details under a unified call interface. This could become V4L2 sub-devices ---------------- Intro ~~~~~ Registration ~~~~~~~~~~~~ **synchronous** **asynchronous** Operations call (or a better name :) ~~~~~~~~~~~~~~~ What do you think ? Thanks j > In the **asynchronous** case subdevice probing can be invoked independently of > the bridge driver availability. The subdevice driver then has to verify whether > all the requirements for a successful probing are satisfied. This can include a > -- > 2.28.0 >
Hi Jacopo, thanks for reviewing. On 15/09/20 15:34, Jacopo Mondi wrote: > Hi Luca, > > On Fri, Sep 04, 2020 at 11:51:41PM +0200, Luca Ceresoli wrote: >> This paragraph provides generic information to explain what v4l2_subdev is >> useful for. Placing it in the middle of paragraphs describing the details >> of subdev registration does not make much sense. Move it near the beginning >> of the section when the v4l2_subdev idea has just been introduced and >> before going into its details. >> >> Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net> >> --- >> Documentation/driver-api/media/v4l2-subdev.rst | 12 ++++++------ >> 1 file changed, 6 insertions(+), 6 deletions(-) >> >> diff --git a/Documentation/driver-api/media/v4l2-subdev.rst b/Documentation/driver-api/media/v4l2-subdev.rst >> index fb66163deb38..1c1e3f9da142 100644 >> --- a/Documentation/driver-api/media/v4l2-subdev.rst >> +++ b/Documentation/driver-api/media/v4l2-subdev.rst >> @@ -12,6 +12,12 @@ Usually these are I2C devices, but not necessarily. In order to provide the >> driver with a consistent interface to these sub-devices the >> :c:type:`v4l2_subdev` struct (v4l2-subdev.h) was created. >> >> +The advantage of using :c:type:`v4l2_subdev` is that it is a generic struct and >> +does not contain any knowledge about the underlying hardware. So a driver might >> +contain several subdevs that use an I2C bus, but also a subdev that is >> +controlled through GPIO pins. This distinction is only relevant when setting >> +up the device, but once the subdev is registered it is completely transparent. >> + >> Each sub-device driver must have a :c:type:`v4l2_subdev` struct. This struct >> can be stand-alone for simple sub-devices or it might be embedded in a larger >> struct if more state information needs to be stored. Usually there is a >> @@ -235,12 +241,6 @@ it can call ``v4l2_subdev_notify(sd, notification, arg)``. This macro checks >> whether there is a ``notify()`` callback defined and returns ``-ENODEV`` if not. >> Otherwise the result of the ``notify()`` call is returned. >> >> -The advantage of using :c:type:`v4l2_subdev` is that it is a generic struct and >> -does not contain any knowledge about the underlying hardware. So a driver might >> -contain several subdevs that use an I2C bus, but also a subdev that is >> -controlled through GPIO pins. This distinction is only relevant when setting >> -up the device, but once the subdev is registered it is completely transparent. >> - > > Have you considered moving the whole part that describes how to call > operations, which comes after the synchronous registration case to a > dedicated sub-section ? The above paragraph makes sense in the context > of describing why v4l2_subdev is advantageous as it abstract the > underlying details under a unified call interface. > > This could become > > V4L2 sub-devices > ---------------- > > Intro > ~~~~~ > > Registration > ~~~~~~~~~~~~ > > **synchronous** > **asynchronous** > > Operations call (or a better name :) > ~~~~~~~~~~~~~~~ > > What do you think ? I think you're right. Incoming v2 with the new layout as you suggested.
diff --git a/Documentation/driver-api/media/v4l2-subdev.rst b/Documentation/driver-api/media/v4l2-subdev.rst index fb66163deb38..1c1e3f9da142 100644 --- a/Documentation/driver-api/media/v4l2-subdev.rst +++ b/Documentation/driver-api/media/v4l2-subdev.rst @@ -12,6 +12,12 @@ Usually these are I2C devices, but not necessarily. In order to provide the driver with a consistent interface to these sub-devices the :c:type:`v4l2_subdev` struct (v4l2-subdev.h) was created. +The advantage of using :c:type:`v4l2_subdev` is that it is a generic struct and +does not contain any knowledge about the underlying hardware. So a driver might +contain several subdevs that use an I2C bus, but also a subdev that is +controlled through GPIO pins. This distinction is only relevant when setting +up the device, but once the subdev is registered it is completely transparent. + Each sub-device driver must have a :c:type:`v4l2_subdev` struct. This struct can be stand-alone for simple sub-devices or it might be embedded in a larger struct if more state information needs to be stored. Usually there is a @@ -235,12 +241,6 @@ it can call ``v4l2_subdev_notify(sd, notification, arg)``. This macro checks whether there is a ``notify()`` callback defined and returns ``-ENODEV`` if not. Otherwise the result of the ``notify()`` call is returned. -The advantage of using :c:type:`v4l2_subdev` is that it is a generic struct and -does not contain any knowledge about the underlying hardware. So a driver might -contain several subdevs that use an I2C bus, but also a subdev that is -controlled through GPIO pins. This distinction is only relevant when setting -up the device, but once the subdev is registered it is completely transparent. - In the **asynchronous** case subdevice probing can be invoked independently of the bridge driver availability. The subdevice driver then has to verify whether all the requirements for a successful probing are satisfied. This can include a
This paragraph provides generic information to explain what v4l2_subdev is useful for. Placing it in the middle of paragraphs describing the details of subdev registration does not make much sense. Move it near the beginning of the section when the v4l2_subdev idea has just been introduced and before going into its details. Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net> --- Documentation/driver-api/media/v4l2-subdev.rst | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)