Hi Sakari,
On Thursday 08 March 2012 19:17:46 Sakari Ailus wrote:
> On Wed, Mar 07, 2012 at 12:31:34PM +0100, Laurent Pinchart wrote:
[snip]
> > > > +static int mt9m032_set_frame_interval(struct v4l2_subdev *subdev,
> > > > + struct v4l2_subdev_frame_interval *fi)
> > > > +{
> > > > + struct mt9m032 *sensor = to_mt9m032(subdev);
> > > > + int ret;
> > > > +
> > > > + if (sensor->streaming)
> > > > + return -EBUSY;
> > > > +
> > > > + memset(fi->reserved, 0, sizeof(fi->reserved));
> > >
> > > I'm not quite sure these should be touched.
> >
> > Why not ? Do you think this could cause a regression in the future when
> > the fields won't be reserved anymore ?
>
> The user is responsible for setting those fields to zero. If we set them to
> zero for them they will start relying on that. At some point that might not
> hold true anymore.
Thinking about it some more, applications should set the reserved fields to 0,
or first issue a get call and modify the fields it's interested in, keeping
the reserved fields at their default value. I'll remove the memset here.
--
Regards,
Laurent Pinchart
--
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
