> 
> Initial conditions for "fast string" operations:
> - EDI and ESI must be 8-byte aligned for the Pentium III processor.
>    EDI must be 8-byte aligned for the Pentium 4 processor.
> - String operation must be performed in ascending address order.
> - The initial operation counter (ECX) must be equal to or greater than 64.

I have seen this before, and I always wondered, does it apply to 8-bit as
well as 32-bit movs, and therefore make using movsl worse than movsb for
moving 64 bytes of 8-byte aligned data ?

-hjalmar