Amin Azez wrote:
> When I saw it I thought it would be better to follow the conntrack hash
> method, where each hash bucket is a list head. The rabbit hole lark for
> hash collisions in ipt_recent is a burden.


True, parts of the problems originated in the use of closed hashing -
sadly enough this didn't even save memory, it used even more than
with open hashing. But thats only one part, its full of bad choices
and bad coding.

Anyway, here goes the first shot at a replacement, it should be fully
compatible. Comments and testing welcome.