Faster AMD64 assembler code

[Eric Bainville]

Hi !

I have been playing with AMD64 assembler code and multi-precision computation.
Under favorable assumptions (inputs multiple of 512-bit, etc.) the code I wrote
is faster than GMP in some cases. I put it on my website, if you 
would like to have
a look and include some of these ideas in GMP.
The cycles are measured on an Athlon 64 X2 4400+ (Toledo core).

http://www.bealto.com/amd64mp1.html
http://www.bealto.com/amd64mp2.html

-- Eric Bainville