On Thu 10 May 2012 10:10:21 Hans de Goede wrote:
> Hi,
>
> Comments inline.
>
> On 05/10/2012 09:05 AM, Hans Verkuil wrote:
> > From: Hans Verkuil<[email protected]>
> >
> > Rather than testing whether an ioctl is implemented in the driver or not
> > every time the ioctl is called, do it upfront when the device is registered.
> >
> > This also allows a driver to disable certain ioctls based on the
> > capabilities
> > of the detected board, something you can't do today without creating
> > separate
> > v4l2_ioctl_ops structs for each new variation.
> >
> > For the most part it is pretty straightforward, but for control ioctls a
> > flag
> > is needed since it is possible that you have per-filehandle controls, and
> > that
> > can't be determined upfront of course.
> >
> > Signed-off-by: Hans Verkuil<[email protected]>
> > ---
> > drivers/media/video/v4l2-dev.c | 171 +++++++++++++++++
> > drivers/media/video/v4l2-ioctl.c | 391
> > +++++++++++---------------------------
> > include/media/v4l2-dev.h | 11 ++
> > 3 files changed, 297 insertions(+), 276 deletions(-)
> >
...
> > diff --git a/drivers/media/video/v4l2-ioctl.c
> > b/drivers/media/video/v4l2-ioctl.c
> > index 3f34098..21da16d 100644
> > --- a/drivers/media/video/v4l2-ioctl.c
> > +++ b/drivers/media/video/v4l2-ioctl.c
> > @@ -526,19 +518,28 @@ static long __video_do_ioctl(struct file *file,
> > return ret;
> > }
> >
> > - if ((vfd->debug& V4L2_DEBUG_IOCTL)&&
> > - !(vfd->debug& V4L2_DEBUG_IOCTL_ARG)) {
> > - v4l_print_ioctl(vfd->name, cmd);
> > - printk(KERN_CONT "\n");
> > - }
> > -
> > if (test_bit(V4L2_FL_USES_V4L2_FH,&vfd->flags)) {
> > vfh = file->private_data;
> > use_fh_prio = test_bit(V4L2_FL_USE_FH_PRIO,&vfd->flags);
> > + if (use_fh_prio)
> > + ret_prio = v4l2_prio_check(vfd->prio, vfh->prio);
> > }
> >
> > - if (use_fh_prio)
> > - ret_prio = v4l2_prio_check(vfd->prio, vfh->prio);
> > + if (v4l2_is_valid_ioctl(cmd)) {
>
> I would prefer for this check to be the first check in the function
> in the form of:
>
> if (!v4l2_is_valid_ioctl(cmd))
> return -ENOTTY;
>
> This will drop an indentation level from the code below and also drop an
> indentation level from the prio check introduced in a later patch,
> making the end result much more readable IMHO.
Ah, no. That's why I renamed v4l2_is_valid_ioctl to v4l2_is_known_ioctl in my
new ioctlv2 branch. v4l2_is_known_ioctl means whether it is a known, standard
v4l2 ioctl. If it is, then it is handled by the table, if it isn't, then it
will be handled by the vidioc_default callback.
I realized myself that that name was very ambiguous.
>
> > + struct v4l2_ioctl_info *info =&v4l2_ioctls[_IOC_NR(cmd)];
> > +
> > + if (!test_bit(_IOC_NR(cmd), vfd->valid_ioctls)) {
> > + if (!(info->flags& INFO_FL_CTRL) ||
> > + !(vfh&& vfh->ctrl_handler))
> > + return -ENOTTY;
> > + }
> > + }
>
> Sort of hard to read, IMHO the below is easier to parse by us humans:
>
> if (!test_bit(_IOC_NR(cmd), vfd->valid_ioctls) &&
> !((info->flags & INFO_FL_CTRL) && vfh && vfh->ctrl_handler))
> return -ENOTTY;
Yeah, I agree. That's better.
Regards,
Hans
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html