From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: many sockets, slow sendto Date: Wed, 21 Mar 2007 23:24:16 +0100 Message-ID: <4601B090.9060704@cosmosbay.com> References: <20070306182039.GJ25760@galon.ev-en.org> <45FF185B.4070007@fw.hu> <20070319.161611.70218081.davem@davemloft.net> <4600592E.80605@fw.hu> <460064C6.5030302@cosmosbay.com> <4601A949.2000005@fw.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-2; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , baruch@ev-en.org, netdev@vger.kernel.org To: Zacco Return-path: Received: from www.cosmosbay.com ([86.65.150.131]:44822 "EHLO gw1.cosmosbay.com" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S965381AbXCUWYp (ORCPT ); Wed, 21 Mar 2007 18:24:45 -0400 In-Reply-To: <4601A949.2000005@fw.hu> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Zacco a =E9crit : > Actually, the source address would be more important in my case, as m= y=20 > clients (each with different IP address) wants to connect to the same= =20 > server, i.e. to the same address and port. I dont understand why you need many sockets then. A single socket should be enough. > I think, the current design is fair enough for server implementations= =20 > and for regular clients. But even though my application is not tipica= l,=20 > as far as I know (but it can be important with the fast performance=20 > growth of regular PCs), the make-up should be general enough to cope=20 > with special circumstances, like mine. My initial idea was to somehow= =20 > include the complete socket pair, i.e. source address:port and=20 > destination address:port, keeping in mind that it should work for bot= h=20 > IPv4 and IPv6. Maybe it's an overkill, I don't know. Could you send me a copy of your application source, or detailed specs,= =20 because I am confused right now...