[PATCH] fix a sign problem with valuator data.

[Peter Hutterer]

On Fri, Oct 15, 2010 at 11:45:07AM -0400, Chase Douglas wrote:
> On Thu, 2010-10-14 at 15:09 -0400, Joe Shaw wrote:
> > From: Joe Shaw <joeshaw at litl.com>
> > 
> > Without this patch, any negative valuator value is wrong when returned
> > from XQueryDeviceState().  This is a regression from at least xserver
> > 1.4.
> > 
> > Valuator data is set in dix/getevents.c:set_valuators() by copying
> > signed int values into an unsigned int field
> > DeviceEvent.valuators.data.
> > 
> > That data is converted into a double with an implicit cast by
> > assignment to axisVal[i] in Xi/exevents.c:UpdateDeviceState().
> > 
> > That double is converted back to a signed int in
> > queryst.c:ProcXQueryDeviceState().  If the original value in
> > set_valuators() is negative, the double value will be > 2^31 and the
> > conversion back to a signed int is undefined.  (Although I
> > consistently see the value -2^31.)
> > 
> > Fix this by changing the definition of DeviceEvent.valuators.data from
> > uint32_t to int32_t.
> > 
> > Signed-off-by: Joe Shaw <joeshaw at litl.com>
> > ---
> >  dix/getevents.c    |    2 +-
> >  include/eventstr.h |    2 +-
> >  2 files changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/dix/getevents.c b/dix/getevents.c
> > index 82bb77b..531595e 100644
> > --- a/dix/getevents.c
> > +++ b/dix/getevents.c
> > @@ -177,7 +177,7 @@ set_valuators(DeviceIntPtr dev, DeviceEvent* event, int first_valuator,
> >      }
> >  
> >      memcpy(&event->valuators.data[first_valuator],
> > -           valuators, num_valuators * sizeof(uint32_t));
> > +           valuators, num_valuators * sizeof(int32_t));
> >  
> >  }
> >  
> > diff --git a/include/eventstr.h b/include/eventstr.h
> > index 433227e..377cceb 100644
> > --- a/include/eventstr.h
> > +++ b/include/eventstr.h
> > @@ -99,7 +99,7 @@ struct _DeviceEvent
> >      struct {
> >          uint8_t  mask[(MAX_VALUATORS + 7)/8]; /**< Valuator mask */
> >          uint8_t  mode[(MAX_VALUATORS + 7)/8]; /**< Valuator mode (Abs or Rel)*/
> > -        uint32_t data[MAX_VALUATORS];         /**< Valuator data */
> > +        int32_t  data[MAX_VALUATORS];         /**< Valuator data */
> >          int32_t  data_frac[MAX_VALUATORS];    /**< Fractional part for data */
> >      } valuators;
> >      struct {
> 
> I can't think of any reason this fix would cause issues except for
> values > 2^31. I don't think that ever happens :).
> 
> Reviewed-by: Chase Douglas <chase.douglas at canonical.com>


oops. looks like a typo, should have probably been the other way round
(data_frac as unsigned and data as signed). 

merged, thanks.

Cheers,
  Peter