NameDateSize

..22-Dec-201645

bn-586.plH A D08-Mar-201513.8 KiB

co-586.plH A D08-Mar-20155.5 KiB

ia64.SH A D08-Mar-201544.3 KiB

mips3.sH A D08-Mar-201537 KiB

mo-586.plH A D08-Mar-201516.3 KiB

pa-risc2.sH A D08-Mar-201547.5 KiB

pa-risc2W.sH A D08-Mar-201545.6 KiB

ppc.plH A D08-Mar-201546.4 KiB

READMEH A D08-Mar-20151 KiB

sparcv8.SH A D08-Mar-201527.6 KiB

sparcv8plus.SH A D08-Mar-201532 KiB

x86/H20-Dec-201610

x86.plH A D08-Mar-2015623

x86_64-gcc.cH A D08-Mar-201617.8 KiB

x86_64-mont.plH A D08-Mar-20154.3 KiB

README

1<OBSOLETE>
2
3All assember in this directory are just version of the file
4crypto/bn/bn_asm.c.
5
6Quite a few of these files are just the assember output from gcc since on 
7quite a few machines they are 2 times faster than the system compiler.
8
9For the x86, I have hand written assember because of the bad job all
10compilers seem to do on it.  This normally gives a 2 time speed up in the RSA
11routines.
12
13For the DEC alpha, I also hand wrote the assember (except the division which
14is just the output from the C compiler pasted on the end of the file).
15On the 2 alpha C compilers I had access to, it was not possible to do
1664b x 64b -> 128b calculations (both long and the long long data types
17were 64 bits).  So the hand assember gives access to the 128 bit result and
18a 2 times speedup :-).
19
20There are 3 versions of assember for the HP PA-RISC.
21
22pa-risc.s is the origional one which works fine and generated using gcc :-)
23
24pa-risc2W.s and pa-risc2.s are 64 and 32-bit PA-RISC 2.0 implementations
25by Chris Ruemmler from HP (with some help from the HP C compiler).
26
27</OBSOLETE>
28