generic-library/openssl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

bn/asm/armv8-mont.pl: boost performance.

Browse Source 
Reviewed-by: Richard Levitte <levitte@openssl.org>
This commit is contained in:
Andy Polyakov 2015-04-20 15:13:24 +02:00
parent d11582f133
commit d38f1b39f1
2 changed files with 1275 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1277
crypto/bn/asm/armv8-mont.pl
View File

File diff suppressed because it is too large Load Diff

17
crypto/ec/asm/ecp_nistz256-armv8.pl
View File

@ -14,19 +14,16 @@
# Original ECP_NISTZ256 submission targeting x86_64 is detailed in # Original ECP_NISTZ256 submission targeting x86_64 is detailed in
# http://eprint.iacr.org/2013/816. # http://eprint.iacr.org/2013/816.
# #
# with/without -DECP_NISTZ256_ASM(*) # with/without -DECP_NISTZ256_ASM
# Apple A7 +140-590% # Apple A7 +120-360%
# Cortex-A53 +135-720% # Cortex-A53 +120-400%
# Cortex-A57 +145-570% # Cortex-A57 +120-350%
# X-Gene +120-700% # X-Gene +200-330%
# Denver +150-740% # Denver +140-400%
#
# (*) comparison is not really "fair", because it's compared to C
# implementation, unlike other similar cases that is;
# #
# Ranges denote minimum and maximum improvement coefficients depending # Ranges denote minimum and maximum improvement coefficients depending
# on benchmark. Lower coefficients are for ECDSA sign, server-side # on benchmark. Lower coefficients are for ECDSA sign, server-side
# operation. Keep in mind that +500% means 6x improvement. # operation. Keep in mind that +400% means 5x improvement.
$flavour = shift; $flavour = shift;
while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {} while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}