On Thu, 2010-12-09 at 10:20 +1000, Peter Hutterer wrote: > On Wed, Dec 08, 2010 at 05:35:06PM -0500, Adam Jackson wrote: > > + If the server supports the X Input Extension version 2 or higher, > > + the devices element names a set of master device to apply the > > + barrier to. If XIAllDevices or XIAllMasterDevices are given, the > > + barrier applies to all master devices. If a slave device is named, > > + BadMatch is generated; this does not apply to slave devices named > > + implicitly by XIAllDevices. Naming a device multiple times is > > + legal, and is treated as though it were named only once. If a > > + device is removed, the barrier continues to apply to the remaining > > + devices, but will not apply to any future device with the same ID > > + as the removed device. Nothing special happens when all matching > > + devices are removed; barriers must be explicitly destroyed. > > + > > + Errors: IDChoice, Window, Value, Match > > BadDevice for invalid device IDs?
Even better.
> > +typedef struct {
> > + CARD8 reqType;
> > + CARD8 xfixesReqType;
> > + CARD16 length B16;
> > + Barrier barrier B32;
> > + Window window B32;
> > + CARD16 x1 B16;
> > + CARD16 y1 B16;
> > + CARD16 x2 B16;
> > + CARD16 y2 B16;
> > + CARD32 directions;
> > + CARD32 num_devices;
>
> XI2 uses CARD16 num_devices everywhere iirc. also allows you to have a 16
> byte pad for later use without having to extend the request length.
Fair enough.
> any comments on the INT16/CARD16 for x/y coordinates?
Yeah, just forgot to update that. Will be fixed in the final version.
> those minor things aside
> Reviewed-by: Peter Hutterer <[email protected]>
Ta!
- ajax
signature.asc
Description: This is a digitally signed message part
_______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
