From: Andy Polyakov Date: Fri, 27 May 2011 18:58:37 +0000 (+0000) Subject: rc4-x86_64.pl: fix due credit. X-Git-Tag: OpenSSL-fips-2_0-rc1~374 X-Git-Url: https://git.librecmc.org/?a=commitdiff_plain;h=f44cb15fabcc22cd06e28666dd87894c8f3f998b;p=oweals%2Fopenssl.git rc4-x86_64.pl: fix due credit. --- diff --git a/crypto/rc4/asm/rc4-x86_64.pl b/crypto/rc4/asm/rc4-x86_64.pl index 4eb0990302..02bfac34f2 100755 --- a/crypto/rc4/asm/rc4-x86_64.pl +++ b/crypto/rc4/asm/rc4-x86_64.pl @@ -78,12 +78,13 @@ # May 2011 # -# The only code path that was not modified is P4-specific one. New -# AMD code path is inspired by and Intel optimization is heavily -# based on submission from Maxim Locktyukhin of Intel. Current -# performance in cycles per processed byte (less is better) and -# improvement coefficients relative to previous version of this -# module are: +# The only code path that was not modified is P4-specific one. Non-P4 +# Intel code path optimization is heavily based on submission by Maxim +# Perminov, Maxim Locktyukhin and Jim Guilford of Intel. I've used +# some of the ideas even in attempt to optmize the original RC4_INT +# code path... Current performance in cycles per processed byte (less +# is better) and improvement coefficients relative to previous +# version of this module are: # # Opteron 5.3/+0% # P4 6.5 @@ -92,9 +93,10 @@ # Sandy Bridge 4.2/+120% # Atom 9.3/+80% # -# (*) Note that this result is ~15% lower than result for 32-bit -# code, meaning that it's possible to improve it, but it's -# more than likely at the cost of the others... +# (*) Note that Core2 result is ~15% lower than corresponding result +# for 32-bit code, meaning that it's possible to improve it, +# but more than likely at the cost of the others (see rc4-586.pl +# to get the idea)... $flavour = shift; $output = shift;