[PATCH video-siliconmotion] Added detection of SM501 on PPC LocalBus.

[Marc Scherer]

This is the first patch in a series to make the Siliconmotion driver
work with the TQM5200 Board.

If no PCI device is found the driver tries to MMIO to the relevant
location for a SM501 on PowerPC Localbus. If it finds the right device
ID after MMIO-mapping it assumes the chip is there. If not it exits with
an error message.

On my TQM5200, gcc reverses the bit-fields (which are heavily used).
Therefore, the bit-accesses to the registers are all wrong. I have not
found a way to reverse the bit-fields back to the right order.

As an example, I have converted the bit-field access to the STATUS
register to bit-masking. But all other bit-fields still have to be
converted that way, too.

Anyway, with this patch, the driver should work as normal with the PCI
devices. I cannot check this, could somebody please confirm?

Regards,
Marc