From mboxrd@z Thu Jan 1 00:00:00 1970 From: Octavian Purdila Subject: Re: [RFC] [PATCH] udp: optimize lookup of UDP sockets to by including destination address in the hash key Date: Thu, 5 Nov 2009 19:03:02 +0200 Message-ID: <200911051903.02859.opurdila@ixiacom.com> References: <4AF1EC18.9090106@ixiacom.com> <200911051825.45749.opurdila@ixiacom.com> <4AF2FF22.2000805@gmail.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Lucian Adrian Grijincu , netdev@vger.kernel.org To: Eric Dumazet Return-path: Received: from ixro-out-rtc.ixiacom.com ([92.87.192.98]:30203 "EHLO ixro-ex1.ixiacom.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1755904AbZKERF7 convert rfc822-to-8bit (ORCPT ); Thu, 5 Nov 2009 12:05:59 -0500 In-Reply-To: <4AF2FF22.2000805@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On Thursday 05 November 2009 18:36:50 you wrote: > Octavian Purdila a =E9crit : > > IIRC, we first saw this issue in VoIP tests with up to 16000 socket= s > > bound on a certain port and IP addresses (each IP address is assign= ed to > > a particular interface). We need this setup in order to emulate lot= s of > > VoIP users each with a different IP address and possible a differen= t L2 > > encapsulation. >=20 > Interesting case indeed, is it SIP 5060 port or RTP ports ? > (I want to know how many messages per second you want to receive) >=20 > An rbtree with 16000 elements has 15 levels, its a lot, but OK > for small trafic. >=20 Yep the signaling port not the RTP port, and yes I think there is a fai= rly =20 small amount of traffic and rbtree might work. BTW, there is another side of this problem, the time to bind() those 16= K=20 sockets before starting the test - at least on 2.6.7 we didn't yet get = to look=20 at this issue on a recent kernel. Thanks, tavi