[Cryptech Tech] Alpha board strategy

[Randy Bush]

>>> Cool. That should be big enough to fit a CHERI CPU. As I mentioned
>>> before, I think this is a better option than ARM on account of having
>>> a far better memory protection model. Also means you don't have to
>>> rely on the CPU being honest (if you believe its hard to subvert an
>>> FPGA, that is).
>>
>> the philosophy seems eminently sensible, but ...
>>
>> if we want to keep under the pay-for-license ceiling, we are on a tight
>> LE budget.  this puts us in a nasty trade-off space.  how many LE does
>> a cheri cpu eat?
> 
> Apparently it depends, but on the order of 100k.

not leaving much room for crypto in the fpga.  

as you may have read, joachim wants to wire and pad a place for a second
fpga in case we have LE scarcity.  this would certainly force that.  and
the fpga ain't cheap; and we want to minimize barriers to folk playing
with it.

there is a second issue we discussed that has not been on this list.  to
keep risk down, the hack with the freescale a9 is using the same chip as
the novena, where we already have (thanks pavel, joachim, and paul) the
eim working between the soc and the fpga.  hence, the same soc with only
the change from xylinx to altera was thought to keep risk down.

but this next iteration is seen as an agile hack.  there will be more.
perhaps the next iteration, or even a half step as the pads and wires
will be there, can move to a cpu on the second fpga slot.  heck, you can
give it a go on your copy of the board!

randy