>> On my Stellation Two "The Mill" card, U6 is a 74LS14
>> and the U14 socket has a daughterboard with two IC's
>> and a switch plugged into it which is an "OS9 ROM
>> Interface".

The daughterboard replaces a 74LS367 with a PROM + 74LS367
which runs A15-A12 through a look up table in the PROM to rearrange
the memory map to better suit OS/9.

Without the daughterboard, the Stellation Mill card has two address
maps. The first is one-one with the 6502, the other inverts A15.

With the daughterboard switch enabled, the OS/9 mapping is:

6809	6502
0xxx	1xxx
...	...
Axxx	Bxxx
Bxxx	Dxxx	L/C
Cxxx	Exxx	L/C
Dxxx	Fxxx	L/C
Exxx	Cxxx	I/O space
Fxxx	0xxx	ZP+stack+text

According to the instructions that come with the OS/9 upgrade, U6
is originally a 74LS14 but is replaced with a 74LS04 when you replace
the 74LS367 with the daughterboard. Must be a timing adjustment.

>> The ROM socket at U11 is empty on my board--does yours
>> have a ROM?
>>
>Mine doesn't have a ROM either.

>Having a quick look at the back of the circuit board where the
>ROM is, you will see a large trace between pin 24 and 21.
>You could fairly safely say that it has to be either a 2716 or
>a 2532 which both require VCC on pin 21 to operate.

The parts list shows a 2716 but A11 is fed through U3 pins 8 and 10
(exclusive NOR gate with pin 9 grounded turning it into an inverter)
before reaching pin 18 (/PE on 2716 and A11 on the 2532). To enable
the ROM you write to $80 to $C0m6 (setting bit 6 of the control
register). $C0m7 disables the A15 address remapping when set.
The manual actually says the socket takes a 2716/2732 but is wrong -
2716/2532 are needed. When enabled, the ROM occupies $F000-FFFF (2532)
or $F800..FFFF (2716) and cannot be accessed from the 6502 - only the 6809
can read it.

Let me know if you have any more questions - I have the OS/9 manuals and
DOS assembler manuals that came with my card.
-- 
David Wilson  School of IT & CS, Uni of Wollongong, Australia