[PATCH] libcamera: v4l2_videodevice: Use bufferType_ in [get|try|set]Format()

[Hans de Goede]

Hi Laurent,

On 7/19/24 9:37 AM, Laurent Pinchart wrote:
> Hi Hans,
> 
> Thank you for the patch.
> 
> On Sat, Jul 13, 2024 at 08:43:27PM +0200, Hans de Goede wrote:
>> V4L2VideoDevice is using the caps to determine which kind of buffers to
>> use with the video-device in 2 different cases:
>>
>> 1. V4L2VideoDevice::open()
>> 2. V4L2VideoDevice::[get|try|set]Format()
>>
>> And the order in which the caps are checked is different between
>> these 2 cases. This is a problem for /dev/video# nodes which support
>> both video-capture and metadata buffers. open() sets bufferType_ to
>> V4L2_BUF_TYPE_VIDEO_CAPTURE[_MPLANE] in this case, where as
>> [get|try|set]Format() will call [get|set]FormatMeta() which does not
>> work with V4L2_BUF_TYPE_VIDEO_CAPTURE[_MPLANE] buffers.
>>
>> Switch [get|try|set]Format() to use the bufferType_ to determine on what
>> sort of buffers they should be operating, leaving the V4L2VideoDevice
>> code with only a single place where the decision is made what sort
>> of buffers it should operate on for a specific /dev/video# node.
>>
>> This will also allow to modify open() in the future to take a bufferType
>> argument to allow overriding the default bufferType it selects for
>> /dev/video# nodes which are capable of supporting more then 1 buffer type.
>>
>> Signed-off-by: Hans de Goede <hdegoede at redhat.com>
>> ---
>>  src/libcamera/v4l2_videodevice.cpp | 56 ++++++++++++++++++++++--------
>>  1 file changed, 41 insertions(+), 15 deletions(-)
>>
>> diff --git a/src/libcamera/v4l2_videodevice.cpp b/src/libcamera/v4l2_videodevice.cpp
>> index 4947aa3d..a8dc5355 100644
>> --- a/src/libcamera/v4l2_videodevice.cpp
>> +++ b/src/libcamera/v4l2_videodevice.cpp
>> @@ -803,12 +803,19 @@ std::string V4L2VideoDevice::logPrefix() const
>>   */
>>  int V4L2VideoDevice::getFormat(V4L2DeviceFormat *format)
>>  {
>> -	if (caps_.isMeta())
>> -		return getFormatMeta(format);
>> -	else if (caps_.isMultiplanar())
>> -		return getFormatMultiplane(format);
>> -	else
>> +	switch (bufferType_) {
>> +	case V4L2_BUF_TYPE_VIDEO_CAPTURE:
>> +	case V4L2_BUF_TYPE_VIDEO_OUTPUT:
>>  		return getFormatSingleplane(format);
>> +	case V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE:
>> +	case V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE:
>> +		return getFormatMultiplane(format);
>> +	case V4L2_BUF_TYPE_META_CAPTURE:
>> +	case V4L2_BUF_TYPE_META_OUTPUT:
>> +		return getFormatMeta(format);
>> +	default:
>> +		return -EINVAL;
> 
> This can never happen, so you could fold the default vase with any of
> the above. I suppose returning -EINVAL could help catching future errors
> when we'll update open() to support more types.

Right I'm not a fan of just adding a default: to some existing case
when there is no clear default existing case.

> but forget to update the
> code here. Is that a real risk ?

Humans make mistakes so yes IMHO this is real risk and I would
prefer to keep this bit as is.

> 
>> +	}
>>  }
>>  
>>  /**
>> @@ -823,12 +830,19 @@ int V4L2VideoDevice::getFormat(V4L2DeviceFormat *format)
>>   */
>>  int V4L2VideoDevice::tryFormat(V4L2DeviceFormat *format)
>>  {
>> -	if (caps_.isMeta())
>> -		return trySetFormatMeta(format, false);
>> -	else if (caps_.isMultiplanar())
>> -		return trySetFormatMultiplane(format, false);
>> -	else
>> +	switch (bufferType_) {
>> +	case V4L2_BUF_TYPE_VIDEO_CAPTURE:
>> +	case V4L2_BUF_TYPE_VIDEO_OUTPUT:
>>  		return trySetFormatSingleplane(format, false);
>> +	case V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE:
>> +	case V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE:
>> +		return trySetFormatMultiplane(format, false);
>> +	case V4L2_BUF_TYPE_META_CAPTURE:
>> +	case V4L2_BUF_TYPE_META_OUTPUT:
>> +		return trySetFormatMeta(format, false);
>> +	default:
>> +		return -EINVAL;
> 
> Same here.
> 
>> +	}
>>  }
>>  
>>  /**
>> @@ -843,12 +857,24 @@ int V4L2VideoDevice::tryFormat(V4L2DeviceFormat *format)
>>  int V4L2VideoDevice::setFormat(V4L2DeviceFormat *format)
>>  {
>>  	int ret = 0;
> 
> While at it you can drop the initialization of the ret variable.

Ack.

> 
>> -	if (caps_.isMeta())
>> -		ret = trySetFormatMeta(format, true);
>> -	else if (caps_.isMultiplanar())
>> -		ret = trySetFormatMultiplane(format, true);
>> -	else
>> +
>> +	switch (bufferType_) {
>> +	case V4L2_BUF_TYPE_VIDEO_CAPTURE:
>> +	case V4L2_BUF_TYPE_VIDEO_OUTPUT:
>>  		ret = trySetFormatSingleplane(format, true);
>> +		break;
>> +	case V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE:
>> +	case V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE:
>> +		ret = trySetFormatMultiplane(format, true);
>> +		break;
>> +	case V4L2_BUF_TYPE_META_CAPTURE:
>> +	case V4L2_BUF_TYPE_META_OUTPUT:
>> +		ret = trySetFormatMeta(format, true);
>> +		break;
>> +	default:
>> +		ret = -EINVAL;
>> +		break;
> 
> And here.
> 
> If you think we should drop the default case, please send a new version.
> Otherwise I can drop the initialization of ret when applying. Either
> way,

As mendtioned above I prefer to keep the default case. If you can
drop the initialization of ret while applying this that would be
great.

> Reviewed-by: Laurent Pinchart <laurent.pinchart at ideasonboard.com>

Thanks & Regsrds,

Hans





>>  
>>  	/* Cache the set format on success. */
>>  	if (ret)
>