Messages in this thread | | | Date | Sun, 4 Jan 2009 23:20:57 -0800 | From | Andrew Morton <> | Subject | Re: 2.6.28-mmotm1230 - include/media/v4l2-ioctl.h prototype mismatch |
| |
On Fri, 02 Jan 2009 05:51:26 -0500 Valdis.Kletnieks@vt.edu wrote:
> I'm seeing the following warning message during a kernel build: > > CC [M] drivers/media/video/gspca/gspca.o > drivers/media/video/gspca/gspca.c:1811: warning: initialization from incompatible pointer type > > The root cause appears to be a missed prototype change in a conversion from > ioctl to unlocked_ioctl - in struct file_operations, the former is an int, > but the latter is a long. So we clean it up. > > I have to admit not having checked deeply for second-order effects, but the > kernel builds without warnings and the result works for me. The only other > use of __video_ioctl2 I can see is in drivers/media/video/ivtv/ivtv-ioctl.c, > and there it's used as a return value from a function already defined to return > a long, so we save an int->long cast... > > Signed-off-by: Valdis Kletnieks <valdis.kletnieks@vt.edu> > > --- linux-2.6.28-mmotm1230/include/media/v4l2-ioctl.h.dist 2009-01-01 17:23:00.000000000 -0500 > +++ linux-2.6.28-mmotm1230/include/media/v4l2-ioctl.h 2009-01-02 05:40:45.000000000 -0500 > @@ -297,7 +297,7 @@ extern int video_usercopy(struct file *f > /* Standard handlers for V4L ioctl's */ > > /* This prototype is used on fops.unlocked_ioctl */ > -extern int __video_ioctl2(struct file *file, > +extern long __video_ioctl2(struct file *file, > unsigned int cmd, unsigned long arg); > > /* This prototype is used on fops.ioctl > --- linux-2.6.28-mmotm1230/drivers/media/video/v4l2-ioctl.c.dist 2009-01-01 17:22:50.000000000 -0500 > +++ linux-2.6.28-mmotm1230/drivers/media/video/v4l2-ioctl.c 2009-01-02 05:45:39.000000000 -0500 > @@ -1852,7 +1852,7 @@ static int __video_do_ioctl(struct file > return ret; > } > > -int __video_ioctl2(struct file *file, > +long __video_ioctl2(struct file *file, > unsigned int cmd, unsigned long arg) > { > char sbuf[128];
That code seems to have magically disappeared from linux-next.
| |