Advanced Micro Devices Inc (AMD)

[mmoy]

> Looks to me that the speed up is by reducing the number of
> memory accesses by keeping variables in registers and
> processing 8 bytes at a times.

I'm on Windows-32 so this is not an option. I can only dream
about those extra GPRs. They do make a difference.

> There's no obvious attempt to align the loop on a code boundary
> - this is a mistake.

Could you elaborate?