> 
> Neither the C nor the assembler has changed much. Although there are
> some subtle changes in the segment register values.

OK, please forgive my ignorance, but I don't know what these are...

> OTOH extemem2() doesn't use any pointers, and the mementry code ought
> to iterate through the memory regions even if the bios writes the info
> to a random address.
> 
> I guess you need to (somehow) find out which code part found the memory
> soze before.

OK, I'll take a shot at this... I re-built /usr/mdec/boot with both the 1.2 
and 1.3 revisions of getextmemx.c, with no difference in behavior, so what 
I think I hear you saying is that maybe it is the boot_xxffsv1 code I 
should look at, since it "appears" first... is this correct?  What would be 
another candidate to look at?

Thanks for your patience.

Rob