You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I wonder why Go's bcrypt is 1.5 times slower than Perl's one? Look (I don't use shell's time for perl to exclude perl's startup time, which is much larger than Go's):
$ time bcrypt-tool hash'0123456789' 10
$2a$10$God/j.dxKvryLSZfCAziMuMeZBfV28Xu24G6liCxlQbCM3PNs17S.
real 0m0.073s
user 0m0.072s
sys 0m0.000s
$ perl -E ' use Crypt::Eksblowfish::Bcrypt qw(en_base64 bcrypt); use Time::HiRes qw(time); $t=time; open my $rand, "<", "/dev/urandom" or die $!; 16 == sysread $rand, my $salt, 16 or die $!; close $rand or die $!; say bcrypt("0123456789", q{$2a$10$}.en_base64($salt)); say time-$t;'$2a$10$WgABAl2bimF7wAOpRebTL.FQdV4n8Ho0/cWA4pGas.PUKpQTbYAL6
0.0554509162902832
Well, to be 100% honest this Perl module is partially implemented in C, but, still, Go shouldn't be 1.5 times slower than C on such tasks, isn't it?
The text was updated successfully, but these errors were encountered:
I wonder why Go's bcrypt is 1.5 times slower than Perl's one? Look (I don't use shell's
time
for perl to exclude perl's startup time, which is much larger than Go's):Well, to be 100% honest this Perl module is partially implemented in C, but, still, Go shouldn't be 1.5 times slower than C on such tasks, isn't it?
The text was updated successfully, but these errors were encountered: