#
#                           LARD 2.0
#            "It's ok to run out of butter in Zambia"
#         "Just smear squashed caterpillars on your toast"
#                            by VIz
#			     
#   I am not responsible for any use of this program by anyone, on any machine
#   for any purpose, anywhere at anytime....


2.0 adds partial subkeys for bcrypt_set_word over 1.0, nothing else.
This optimization improves the time for a set_word/encode pair by about
a conservative 60%. It will affect performance less if you do more encode's
per set_word.

One drawback of this optimization is that is adds another 128k of tables
to your field length, but, hey, suck is life.

The 128k table is pregenerated, and compiled into the program, expect a long
build of bcrypt.c, the table takes a while to do.

More optimizations are yet to be done, I still have a couple major ones
to do, once those are done Ill spend my time doing diddly shit for
different architectures type stuff.

I dont plan on working on password selection stuff or anything like that,
my main concern is getting bcrypt faster.

If you can't get it working, then you don't deserve to run it.

VIz
