simulator (Re: [Icfp04-discuss] three liberal arts)

[Jed Davis]

Eijiro Sumii <sumii at saul.cis.upenn.edu> writes:

> Just for information (and fun:-)...
>
> There are _many_ different ways to implement an ant simulator.  For
> example, one could compile ant code into C and thereby remove
> interpretive overheads.

I'd thought of that, although I have to wonder how much improvement it
gives.  Certainly, saving the switch (and how are those implemented,
nowadays?) on the ant-opcode, and on the parameters for Sense would be
nice...  but how nice?  And what of locality of reference in the
compiled code, especially for the huge 8000-insn ants?  

Meanwhile, I find myself wondering what the performance implications
would be of using smaller integers than full words (e.g., half-words,
bytes, and bit fields).  I mean, I keep hearing that memory accesses
are the big performance bottleneck these days, so I wonder if it would
help to save N-1 loads (and N-1 words of dcache?) at the cost of N
rlwinm's (or whatever the target uses for bit-masking).


(Once my new simulator is working, I'll be able to test some of this.)
-- 
dn: cn=Jed Davis, o=panix.com  ##          see also jldavis at cs.oberlin.edu
objectclass: person
mail;personal: jdev at panix.com  ##    PGP Key FP:   A098 903E 9B9A DEF4 168F
mail;work:     jld@/           ##  [id 0xF33659F9] AA09 BF07 807E F336 59F9
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 185 bytes
Desc: not available
Url : http://lists.seas.upenn.edu/pipermail/icfp04-discuss/attachments/20040608/79fe21a4/attachment.bin