ECOMP (was:Re: OO vs. CO)

Paulo Ferreira paf@REDACTED
Tue Mar 11 17:31:36 CET 2003


>> I know nothing of such things but I am left wondering after
>> someone commented on the list about the possibility of using the
>> Crusoe chip to make an Erlang processor.  Would that "just require
>> software"?
>
>	As far as I know, even if the Crusoe instruction set can be
>software-redefined, the interface to do so is not published.

Sad, very sad.....

Now, some crazy ideas:

Mildly crazy:

 If the Beam specification matches well a virtual stack machine, why not
  use a real stack machine like the PTSC IGNITE ( http://www.ptsc.com )?
   Those guys sell the processor and the "IP" to put the processor on
  your own FPGA.

 BTW is there something like "Beam specification for dummies"?


Really really crazy:  :-) :-)
  Why not "compile erlang directly into Hardware" ?

   We have usually lots of processes, that could be "mapped" into
   different areas/blocks of the FPGA.

   Each process has its own hardware, enough to do its own tasks,
   and comunicates with other "processes/hardware blocks" by
   message passing.

 Things like this loop are simple and fast in Hardware.

   loop(Counter)->
        receive
           {up,Pid} -> loop(Counter+1);
           {down,Pid} -> loop(Counter-1);
           {query,Pid} -> Pid ! {value,Counter}, loop(Counter)
        end.


   Variants:
   a) At startup, all "processes/hardware blocks" are created and start
      "processing" the messages they receive.

    b) The "hardware blocks/processes" are "created" at runtime
       reconfiguring dynamically the FPGA.

   So instead of making hardware that "paralelizes" serial software
   with things like "sliding registers", why not start with software
   that is already paralell like Erlang, and put it into Hardware ?



  Greetings
 Paulo Ferreira









------------------------------------------------
Paulo Ferreira paf@REDACTED 





More information about the erlang-questions mailing list