[PULL] hw/xwin fixes + a warning fix

[Peter Hutterer]

On Thu, Jan 26, 2012 at 01:10:53PM -0800, Keith Packard wrote:
> On Thu, 26 Jan 2012 19:27:42 -0000, "Colin Harrison" <colin.harrison at virgin.net> wrote:
> 
> > > This looks like it's missing
> > >       x = dx;
> > >       y = dy;
> > 
> > With this in I get no mouse events (on the left monitor) when the main
> > display is not leftmost with multiple monitors.
> 
> I don't understand how that can happen -- miSetPointerPosition isn't
> supposed to change the position unless the pointer is constrained.
> 
> > The casting is probably messing up negative x's on Windows? As in the [PULL]
> > I can't detect any position errors.
> 
> You won't be clipping mouse positions, which makes this change
> significantly different from the commit message (as it now discards
> information that it used to use).
> 
> Perhaps Peter has some idea of what's going on here...

quick peek only but just removing the miSetPointerPosition() call
altogether seems like the right thing to do. Not sure why the other one
doesn't move the pointer but rough guess: with the newfangled coord scaling
using the clipped valuators in QueuePointerEvents may actually cancel out
the miPointerSetPosition() effect. That's really just a guess though.

Jon confirmed on IRC that just removing miSetPointerPosition() fixes it
anyways, so let's do that.
 
Cheers,
  Peter