Date: Tue, 30 Dec 2008 19:15:18 -0800 From: "Kip Macy" <kmacy@freebsd.org> To: "Bruce M. Simpson" <bms@freebsd.org> Cc: src-committers@freebsd.org, svn-src-user@freebsd.org Subject: Re: svn commit: r186585 - user/kmacy/HEAD_fast_net/sys/netinet Message-ID: <3c1674c90812301915y5c42b01al445667b244cb0633@mail.gmail.com> In-Reply-To: <495A4F27.1050502@FreeBSD.org> References: <200812300438.mBU4ceDK076135@svn.freebsd.org> <495A4F27.1050502@FreeBSD.org>
index | next in thread | previous in thread | raw e-mail
On Tue, Dec 30, 2008 at 8:41 AM, Bruce M. Simpson <bms@freebsd.org> wrote: > Kip Macy wrote: >> >> Author: kmacy >> Date: Tue Dec 30 04:38:39 2008 >> New Revision: 186585 >> URL: http://svn.freebsd.org/changeset/base/186585 >> >> Log: >> make ipv4 forwarding flowtable aware >> > > This is so that we continue to use the same IPv4 next-hop consistently in an > ECMP enabled kernel for sockets, right? Caching the rtentry in the inpcb (an earlier commit) already does that for connected sockets. This allows ip_forward to consistently forward packets to the same host so that ECMP can be used for server load balancing. Cheers, Kiphelp
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3c1674c90812301915y5c42b01al445667b244cb0633>