Date: Wed, 15 Aug 2018 01:54:43 +0200 From: Kajetan Staszkiewicz <vegeta@tuxpowered.net> To: Ermal =?ISO-8859-1?Q?Lu=E7i?= <eri@freebsd.org> Cc: Kristof Provost <kp@freebsd.org>, "freebsd-pf@freebsd.org" <freebsd-pf@freebsd.org> Subject: Re: pf tables locking Message-ID: <6021147.AAtAggGk6h@energia> In-Reply-To: <CAPBZQG1S=M4DFZytRzYWD0HeT3yjm6HLCAA6HEb-Td0jg0svHQ@mail.gmail.com> References: <8680316.SccKl5VnxN@energia> <13826523.m2ultlLLsi@energia> <CAPBZQG1S=M4DFZytRzYWD0HeT3yjm6HLCAA6HEb-Td0jg0svHQ@mail.gmail.com>
index | next in thread | previous in thread | raw e-mail
[-- Attachment #1 --] On Tuesday, 14 August 2018 15:44:52 CEST Ermal Luçi wrote: > If you really want to spend time on it, the best option is to pull out the > pool concept used by the rules/nat... and manage it outside of the > rules/states but in its own module referenced by the former ones. Do you mean as separate kernel module? Or totally outside of kernel? I was considering doing this outside of kernel by providing a weighted round-robin algorithm but that would still require most of the patches as for doing it within kernel, in order to get counters working for redirection tables and state counter per table element, which both are missing in kernel now. > This would allow extensibility and propper reasoning about it. It might be the late hour but I really don't see how it would be extensible. Please be more specific. -- | pozdrawiam / greetings | powered by Debian, FreeBSD and CentOS | | Kajetan Staszkiewicz | jabber,email: vegeta()tuxpowered net | | Vegeta | www: http://vegeta.tuxpowered.net | `------------------------^---------------------------------------' [-- Attachment #2 --] -----BEGIN PGP SIGNATURE----- iF0EABECAB0WIQSOEQZObv2B8mf0JbnjtFCvbXs6FAUCW3NrwwAKCRDjtFCvbXs6 FE1dAJ979AM5qro0P+tx/f1WbBTnKJIXVQCgmaCW6/OG3hfWoxKzoIVEWHlZXgA= =3czq -----END PGP SIGNATURE-----help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6021147.AAtAggGk6h>