Athlon XP Single limb multiply
Rickey Bowers Jr.
bit at savoldi.com
Sun Dec 21 15:59:23 CET 2003
Sorry I do not use those tools. The change basically involves putting zero
in ECX and moving that value into EBX - reducing the loop code to 14 bytes -
instead of the 17 bytes currently used. Padding could be used to make the
loop code 16 bytes as well.
Hopefully, an assembly language programmer which uses AT&T syntax will be
able to incorporate this change for the benefit of all.
A 10% improvement is nothing. Should I not post any more MASM code?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the gmp-devel