Em 05-07-2012 19:36, Sylwester Nawrocki escreveu:
> On 07/06/2012 12:11 AM, Mauro Carvalho Chehab wrote:
>>> +static int vidioc_dqbuf(struct file *file, void *priv, struct v4l2_buffer *p)
>>> +{
>>> + struct stk1160 *dev = video_drvdata(file);
>>> +
>>> + if (!stk1160_is_owner(dev, file))
>>> + return -EBUSY;
>>> +
>>> + return vb2_dqbuf(&dev->vb_vidq, p, file->f_flags& O_NONBLOCK);
>>
>> Why to use O_NONBLOCK here? it should be doing whatever userspace wants.
>
> This is OK, since the third argument to vb2_dqbuf() is a boolean indicating
> whether this call should be blocking or not. And a "& O_NONBLOCK" masks this
> information out from file->f_flags.
Ah! OK then.
It might be better to initialize it during vb2 initialization, at open,
instead of requiring this argument every time vb_dqbuf() is called.
Btw, just noticed a minor issue: an space is required before the "&" operator.
Regards,
Mauro
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
