> [T]his is what I'd like to have in the config file:

> ums0 at uhub?
> ums1 at uhub?
> psm0 at pckbc?
> wsmouse0 at ums0
> wsmouse0 at ums1
> wsmouse0 at psm0

> That is, wsmouse0 (i.e., /dev/wms0) will get the events from all
> three mice.  [...]  Well, the autoconf model that NetBSD has does not
> support multiple parents for a device, so we cant have this.

> If there is no other solution I have a slightly hacky solution.

> wsmouse0 at ums0 destination 0
> wsmouse1 at ums1 destination 0
> wsmouse2 at psm0 destination 0

> This would direct wsmouse1 and wsmouse2 to send their events to
> wsmouse0.

Biggest problem I see with this is, what if ums0 isn't found?  (Perhaps
the machine doesn't have any uhubs, so ums0 and ums1 can't be found
because they're parentless.)  Then there's no wsmouse0 for wsmouse1 and
wsmouse2 to send their events to.  Or can the autoconfig machinery cope
with that?

Someone suggested a pseudo-device.  The biggest problem I see with that
is, someone has to configure which real mice it's to get input from,
and how does whoever that is find out what real mice exist?

Perhaps the pseudo-device should be brought in automatically by
configuring the mouse drivers, and the mouse drivers register
themselves with the generic mouse driver when they're found, so the
generic driver can provide a list of real mice that were found....

And, nobody seems to have even mentioned the possibility of making it
possible for devices to have multiple parents.  Why not?  Is it Just
Too Hard, does it break something else, or what?

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B