-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

November 28th 2009 for gnupg-users@gnupg.org thread "GnuPG private key
resilience against off-line brute-force attacks"

Loop unrolling only gives more performance in very small loops, for
not so small ones there can be in fact a performance penality since as
the unrolled code is great it leaves less cache for data.

The complexity of a S2K algoritm is constant for variable input and
constant iterations, in other words, it is O(1) but this O(1) assumes
constant number of iterations, if we consider that factor the
complexity would be O(iterations).

So that O(1) than you say is correct but meaningless in this context.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEAREIAAYFAksRpCIACgkQZ4DA0TLic4iEUACgjxnvVcF0JXiBI3MuMv8HHwdY
+P4AniUvv+j5Ysg99Qc+xDZ9e1LnCzxS
=h116
-----END PGP SIGNATURE-----

_______________________________________________
Gnupg-users mailing list
Gnupg-users@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-users

Reply via email to