Re: [PATCH v3 06/11] media: v4l2-fwnode: Add helper to parse device properties

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Jacopo,

On 9/13/19 9:04 PM, Jacopo Mondi wrote:
> Hi Hans,
> 
> On Fri, Sep 13, 2019 at 04:08:10PM +0200, Hans Verkuil wrote:
>> On 9/12/19 10:10 PM, Jacopo Mondi wrote:
>>> Add an helper function to parse common device properties in the same
>>> way as v4l2_fwnode_endpoint_parse() parses common endpoint properties.
>>>
>>> Initially parse the 'rotation' and 'location' properties from the
>>
>> Just drop 'Initially'. It's redundant.
>>
>>> firmware interface.
>>>
>>> Signed-off-by: Jacopo Mondi <jacopo@xxxxxxxxxx>
>>> ---
>>>  drivers/media/v4l2-core/v4l2-fwnode.c | 42 +++++++++++++++++++++++++++
>>>  include/media/v4l2-fwnode.h           | 40 +++++++++++++++++++++++++
>>>  2 files changed, 82 insertions(+)
>>>
>>> diff --git a/drivers/media/v4l2-core/v4l2-fwnode.c b/drivers/media/v4l2-core/v4l2-fwnode.c
>>> index 3bd1888787eb..d325a2d5c088 100644
>>> --- a/drivers/media/v4l2-core/v4l2-fwnode.c
>>> +++ b/drivers/media/v4l2-core/v4l2-fwnode.c
>>> @@ -595,6 +595,48 @@ void v4l2_fwnode_put_link(struct v4l2_fwnode_link *link)
>>>  }
>>>  EXPORT_SYMBOL_GPL(v4l2_fwnode_put_link);
>>>
>>> +int v4l2_fwnode_device_parse(struct device *dev,
>>> +			     struct v4l2_fwnode_device_properties *props)
>>> +{
>>> +	struct fwnode_handle *fwnode = dev_fwnode(dev);
>>> +	u32 val;
>>> +	int ret;
>>> +
>>> +	memset(props, 0, sizeof(*props));
>>> +
>>> +	dev_dbg(dev, "===== begin V4L2 device properties parsing\n");
>>> +	ret = fwnode_property_read_u32(fwnode, "location", &val);
>>> +	if (!ret) {
>>> +		switch (val) {
>>> +		case V4L2_FWNODE_LOCATION_FRONT:
>>> +		case V4L2_FWNODE_LOCATION_BACK:
>>> +		case V4L2_FWNODE_LOCATION_EXTERNAL:
>>> +			break;
>>> +		default:
>>> +			dev_warn(dev, "Unsupported device location: %u\n", val);
>>> +			return -EINVAL;
>>> +		}
>>> +
>>> +		props->location = val;
>>> +		dev_dbg(dev, "device location: %u\n", val);
>>> +	}
>>> +
>>> +	ret = fwnode_property_read_u32(fwnode, "rotation", &val);
>>> +	if (!ret) {
>>> +		if (val >= 360 || val % 90) {
>>> +			dev_warn(dev, "Unsupported device rotation: %u\n", val);
>>> +			return -EINVAL;
>>> +		}
>>> +
>>> +		props->rotation = val;
>>> +		dev_dbg(dev, "device rotation: %u\n", val);
>>> +	}
>>> +	dev_dbg(dev, "===== end V4L2 device properties parsing\n");
>>
>> Are these dev_dbg lines really needed? It seems a bit overkill to me.
>>
> 
> It's the same debug output we have in v4l2_fwnode_endpoint_parse() so
> I tried to be consistent. Should I drop it?

I think so, but check with Sakari. If he wants to keep it, then that's fine
with me.

> 
>> What if rotation is set, but not location. Then location defaults to Front.
>> Is that what we want, or should we add an UNKNOWN location?
>>
> 
> As you have noticed I have added a value to encode UNKNOW in the later
> patch, I should probably add them in this one as you suggested ?

Yes, please.

> 
>>> +
>>> +	return 0;
>>> +}
>>> +EXPORT_SYMBOL_GPL(v4l2_fwnode_device_parse);
>>> +
>>>  static int
>>>  v4l2_async_notifier_fwnode_parse_endpoint(struct device *dev,
>>>  					  struct v4l2_async_notifier *notifier,
>>> diff --git a/include/media/v4l2-fwnode.h b/include/media/v4l2-fwnode.h
>>> index f6a7bcd13197..86af6d9d11fe 100644
>>> --- a/include/media/v4l2-fwnode.h
>>> +++ b/include/media/v4l2-fwnode.h
>>> @@ -109,6 +109,28 @@ struct v4l2_fwnode_endpoint {
>>>  	unsigned int nr_of_link_frequencies;
>>>  };
>>>
>>> +/**
>>> + * enum v4l2_fwnode_location - possible device locations
>>> + * @V4L2_FWNODE_LOCATION_FRONT: device installed on the front side
>>> + * @V4L2_FWNODE_LOCATION_BACK: device installed on the back side
>>> + * @V4L2_FWNODE_LOCATION_EXTERNAL: device externally located
>>> + */
>>> +enum v4l2_fwnode_location {
>>> +	V4L2_FWNODE_LOCATION_FRONT,
>>> +	V4L2_FWNODE_LOCATION_BACK,
>>> +	V4L2_FWNODE_LOCATION_EXTERNAL
>>> +};
>>> +
>>> +/**
>>> + * struct v4l2_fwnode_device_properties - fwnode device properties
>>> + * @location: device location. See &enum v4l2_fwnode_location
>>> + * @rotation: device rotation
>>> + */
>>> +struct v4l2_fwnode_device_properties {
>>> +	enum v4l2_fwnode_location location;
>>> +	unsigned int rotation;
>>> +};
>>> +
>>>  /**
>>>   * struct v4l2_fwnode_link - a link between two endpoints
>>>   * @local_node: pointer to device_node of this endpoint
>>> @@ -233,6 +255,24 @@ int v4l2_fwnode_parse_link(struct fwnode_handle *fwnode,
>>>   */
>>>  void v4l2_fwnode_put_link(struct v4l2_fwnode_link *link);
>>>
>>> +/**
>>> + * v4l2_fwnode_device_parse() - parse fwnode device properties
>>> + * @dev: pointer to &struct device
>>> + * @props: pointer to &struct v4l2_fwnode_device_properties where to store the
>>> + *	   parsed properties values
>>> + *
>>> + * This function parses and validates the V4L2 fwnode device properties from
>>> + * the firmware interface. It is responsibility of the caller to allocate a
>>> + * valid @struct v4l2_fwnode_device_properties structure and provide a pointer
>>> + * to it in the @props parameter.
>>> + *
>>> + * Return:
>>> + *	% 0 on success
>>> + *	%-EINVAL if a parsed property value is not valid
>>> + */
>>> +int v4l2_fwnode_device_parse(struct device *dev,
>>> +			     struct v4l2_fwnode_device_properties *props);
>>> +
>>>  /**
>>>   * typedef parse_endpoint_func - Driver's callback function to be called on
>>>   *	each V4L2 fwnode endpoint.
>>>
>>
>> Regards,
>>
>> 	Hans

Regards,

	Hans



[Index of Archives]     [Linux Input]     [Video for Linux]     [Gstreamer Embedded]     [Mplayer Users]     [Linux USB Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [Yosemite Backpacking]

  Powered by Linux