openssl

History

Andy Polyakov 1cbdbcd587 x86_64 performance patch.		2002-12-14 20:45:39 +00:00
..
asm	Remove Win32 assembler files. They are always rebuilt (with some	2000-03-13 08:04:20 +00:00
.cvsignore	Ignore Makefile.save	1999-04-29 16:04:54 +00:00
Makefile.ssl	make update	2002-12-05 01:55:24 +00:00
README	Import of old SSLeay release: SSLeay 0.9.0b	1998-12-21 10:56:39 +00:00
ripemd.h	Make EVP_Digest*() routines return a value.	2001-03-08 14:04:22 +00:00
rmd160.c	Support the EBCDIC character set and BS2000/OSD-POSIX (work in progress).	1999-06-04 21:35:58 +00:00
rmd_dgst.c	x86_64 performance patch.	2002-12-14 20:45:39 +00:00
rmd_locl.h	openbsd-x86 macros	2001-10-14 00:57:30 +00:00
rmd_one.c	Windows CE updates, contributed by Steven Reddie <smr@essemer.com.au>	2002-12-03 14:21:32 +00:00
rmdconst.h	Import of old SSLeay release: SSLeay 0.9.0b	1998-12-21 10:56:39 +00:00
rmdtest.c	Have all tests use EXIT() to exit rather than exit(), since the latter doesn't	2002-11-28 18:57:30 +00:00

README

RIPEMD-160
http://www.esat.kuleuven.ac.be/~bosselae/ripemd160.html

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

eric (20-Jan-1998)