[PATCH synaptics] Init num_touches to 0 on start

[Peter Hutterer]

On Wed, Apr 25, 2012 at 08:01:57PM -0700, Chase Douglas wrote:
> On 04/25/2012 07:56 PM, Peter Hutterer wrote:
> > On Wed, Apr 25, 2012 at 06:22:07PM -0700, Chase Douglas wrote:
> >> On 04/25/2012 05:36 PM, Peter Hutterer wrote:
> >>> We implicitly rely on this already since we calloc the struct. Do it
> >>> expliclity on DeviceOn().
> >>>
> >>> Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>
> >>> ---
> >>> This half of the fix to 49161 a bit redundant but better safe than sorry,
> >>> they say.
> >>
> >> I have not problems with it, but this is basically how things are done
> >> all over the X stack. It's assumed structs are calloc'd where it isn't
> >> obvious pretty much. Does this mean we have to take up arms against all
> >> uninitialized members? :)
> > 
> > in this case the struct alloc is only called once (DEVICE_INIT) but the
> > struct is re-used for each DeviceOn(). So forcing it to known zero before
> > it's being used seems sensible.
> 
> Ok. Is this an actual bug fix then? my impression when I read it was
> that this just made the code make more sense to read.

The first hunk of http://patchwork.freedesktop.org/patch/10055/ has the same
effect in terms of fixing the suspend bug. I just want this one in for
consistency.

Cheers,
  Peter