Date: Sun, 2 Mar 1997 01:27:46 -0500 From: "David S. Miller" <davem@jenolan.rutgers.edu> To: imb@scgt.oz.au Cc: dg@root.com, netdev@roxanne.nuclecu.unam.mx, hackers@FreeBSD.ORG Subject: Re: ok, final sockhash changes, new diff Message-ID: <199703020627.BAA09925@jenolan.caipgeneral> In-Reply-To: <199703020621.RAA12400@asstdc.scgt.oz.au> (message from michael butler on Sun, 2 Mar 1997 17:21:35 %2B1100 (EST))
next in thread | previous in thread | raw e-mail | index | archive | help
From: michael butler <imb@scgt.oz.au> Date: Sun, 2 Mar 1997 17:21:35 +1100 (EST) This being the case, a simple counter (number of listens performed) added to the hash would help redistribute across buckets - don't care about wrap-around .. or am I off the planet ? :-) Huh? The placement within the tables is a product of the hash, and thus the length of the chains is as well. No counter can help change this situation. The only solution is to hash in the local IP address contained in the socket's identity. ---------------------------------------------//// Yow! 11.26 MB/s remote host TCP bandwidth & //// 199 usec remote TCP latency over 100Mb/s //// ethernet. Beat that! //// -----------------------------------------////__________ o David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199703020627.BAA09925>