2 http://www.esat.kuleuven.ac.be/~bosselae/ripemd160.html
4 This is my implementation of RIPEMD-160. The pentium assember is a little
5 off the pace since I only get 1050 cycles, while the best is 1013.
6 I have a few ideas for how to get another 20 or so cycles, but at
7 this point I will not bother right now. I believe the trick will be
8 to remove my 'copy X array onto stack' until inside the RIP1() finctions the
9 first time round. To do this I need another register and will only have one
10 temporary one. A bit tricky.... I can also cleanup the saving of the 5 words
11 after the first half of the calculation. I should read the origional
12 value, add then write. Currently I just save the new and read the origioal.
13 I then read both at the end. Bad.