README revision 55714 
1195609SmpRIPEMD-160
259243Sobrienhttp://www.esat.kuleuven.ac.be/~bosselae/ripemd160.html
359243Sobrien
459243SobrienThis is my implementation of RIPEMD-160.  The pentium assember is a little
559243Sobrienoff the pace since I only get 1050 cycles, while the best is 1013.
659243SobrienI have a few ideas for how to get another 20 or so cycles, but at
759243Sobrienthis point I will not bother right now.  I belive the trick will be
859243Sobriento remove my 'copy X array onto stack' until inside the RIP1() finctions the
959243Sobrienfirst time round.  To do this I need another register and will only have one
1059243Sobrientemporary one.  A bit tricky....  I can also cleanup the saving of the 5 words
1159243Sobrienafter the first half of the calculation.  I should read the origional
1259243Sobrienvalue, add then write.  Currently I just save the new and read the origioal.
1359243SobrienI then read both at the end.  Bad.
1459243Sobrien
1559243Sobrieneric (20-Jan-1998)
1659243Sobrien