Re: [PATCH] em28xx: return -ENOTTY for tuner + frequency ioctls if the device has no tuner

["Frank Schäfer" <fschaefer.oss@googlemail.com>]

Am 14.01.2013 10:20, schrieb Hans Verkuil:
> On Sun January 13 2013 13:50:50 Frank Schäfer wrote:
>> Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
>> ---
>>  drivers/media/usb/em28xx/em28xx-video.c |    8 ++++++++
>>  1 Datei geändert, 8 Zeilen hinzugefügt(+)
>>
>> diff --git a/drivers/media/usb/em28xx/em28xx-video.c b/drivers/media/usb/em28xx/em28xx-video.c
>> index 2eabf2a..4a7f73c 100644
>> --- a/drivers/media/usb/em28xx/em28xx-video.c
>> +++ b/drivers/media/usb/em28xx/em28xx-video.c
>> @@ -1204,6 +1204,8 @@ static int vidioc_g_tuner(struct file *file, void *priv,
>>  	struct em28xx         *dev = fh->dev;
>>  	int                   rc;
>>  
>> +	if (dev->tuner_type == TUNER_ABSENT)
>> +		return -ENOTTY;
>>  	rc = check_dev(dev);
>>  	if (rc < 0)
>>  		return rc;
>> @@ -1224,6 +1226,8 @@ static int vidioc_s_tuner(struct file *file, void *priv,
>>  	struct em28xx         *dev = fh->dev;
>>  	int                   rc;
>>  
>> +	if (dev->tuner_type == TUNER_ABSENT)
>> +		return -ENOTTY;
>>  	rc = check_dev(dev);
>>  	if (rc < 0)
>>  		return rc;
>> @@ -1241,6 +1245,8 @@ static int vidioc_g_frequency(struct file *file, void *priv,
>>  	struct em28xx_fh      *fh  = priv;
>>  	struct em28xx         *dev = fh->dev;
>>  
>> +	if (dev->tuner_type == TUNER_ABSENT)
>> +		return -ENOTTY;
>>  	if (0 != f->tuner)
>>  		return -EINVAL;
>>  
>> @@ -1255,6 +1261,8 @@ static int vidioc_s_frequency(struct file *file, void *priv,
>>  	struct em28xx         *dev = fh->dev;
>>  	int                   rc;
>>  
>> +	if (dev->tuner_type == TUNER_ABSENT)
>> +		return -ENOTTY;
>>  	rc = check_dev(dev);
>>  	if (rc < 0)
>>  		return rc;
>>
> Rather than doing this in each ioctl, I recommend using v4l2_disable_ioctl
> instead. See for example drivers/media/pci/ivtv/ivtv-streams.c.

Hmm, thanks.
I just did the same we currently do for the VIDIOC_G/S/QUERY_STD and
VIDIOC_G/S_AUDIO ioctls, but yeah, disabling seems to be better.
Btw, what about VIDIOC_G/S_PARAM ? Do they make sense for cameras ?

Regards,
Frank