> or so, and of course fix the X driver to mmap() on the device with
> offset == 0xf0000 (the REX3 registers on newport are offset into the GIO
> device space for some reason). This way the X driver doesn't need to
> know which GIO slot the device is in either...

i may not be a kernel hacker (yet) but i'm inclined to think that the
newport can't be in an arbitrary slot. iirc, the extremley rare dual-head
indy configuration didn't use a second newport anyway, it was a special
variant of the GR3-Elan hardware. i'm not sure a second newport would
even physically connect in any meaningful way, and it doesn't look like
a newport would fit atop a challenge-S scsi/ethernet riser either.

but supporting an arbitrary GIO slot for the newport is admirable from an
architectural cleanliness standpoint, so it's hardly surprising to see it
mentioned here :)