26 Jan
2010
26 Jan
'10
5:32 p.m.
It's not just software - I have some "Rev
0" COMBOARDs where the
endiness of the onboard 68000 didn't logically match up with the
endiness of the Unibus - there were hardware and software hacks for
the prototype boards to get them working enough for the developers to
have a platform. Rev 1 boards had the high and low bytes swapped so
that writing a 16-bit quantity on one processor would result in the
same 16-bit quantity read by the other processor.
Alpha Micro had this problem when they went from WD-16 to 68K, and ended
up doing the same thing (swapping lines) to keep the endianness the same
for client data.
Obviously, when they went x86 they simply did software emulation, which
is much less prosaic.
--
------------------------------------ personal: http://www.cameronkaiser.com/ --
Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- Who are you going to believe, me or your own eyes? -- Groucho Marx ---------