>MACH_MACLC475 is defined in sys/arch/mac68k/include/cpu.h on line 183. Try
>adding #define MACH_MACLC475_CHIRPED 90  on the next line.
>
>Then the only other place MACH_MACLC475 is used is in the rom setup, on line
>1969 of sys/arch/mac68k/mac68k/machdep.c:
>{MACH_MACLC475, "LC", " 475 ", MACH_CLASSQ,  &romvecs[9]},
>
>Add something like:
>{MACH_MACLC475_CHIRPED, "LC", " 475 ", MACH_CLASSQ,  &romvecs[9]},
>
>Scott, Allen, Bob, how does this look?
>
Looks fine to me.  As long as no other Mac returns a Gestalt ID of zero 
this could even be included in the -current tree, but I suspect there are 
a couple of other systems that also return zero.  So it would probably be 
a good idea to get the Booter to send in a unique value that could be 
used here.  Maybe even have the booter make the Gestalt ID a user 
configurable parameter?

>Also, you might check the rom vectors on your machine, just to make sure
>they are the same as for the non-chirped sytem.
>
Good point.  Even Apple selects different ROM routines based on things 
other than the Gestalt ID.  That's one of the reasons why the MRG code 
reached a dead end with some families of systems, like the 63x.

>The above two changes, assuming no ROM-vector changes, are all you need to
>do to add support. Assuming your computer behaves just like a 475 w/ a
>different ID.

As far as I remember the Gestalt ID is only used for the MRG code in the 
Kernel.  So if you're running with the HWDIRECT code it shouldn't make 
any difference what the value is and that's one of the many benefits of 
John's code.  I'm just guessing, but I'd bet the 475 is in the same group 
as the '030 Cuda machines and that's now running in HWDIRECT although 
probably not yet in the -current tree.  We may finally be close to the 
day when the MRG code is history and we can all run with the HWDIRECT 
solution.

-bob