From mboxrd@z Thu Jan 1 00:00:00 1970 From: Denys Fedoryschenko Subject: Re: Allowing more than 64k connections and heavily optimize bind(0) time. Date: Sun, 11 Jan 2009 15:03:10 +0200 Message-ID: <200901111503.10736.denys@visp.net.lb> References: <20081224190926.GA3244@ioremap.net> <20090110.232019.98466627.davem@davemloft.net> <20090111125206.GA24173@ioremap.net> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Cc: David Miller , netdev@vger.kernel.org To: Evgeniy Polyakov Return-path: Received: from hosting.visp.net.lb ([194.146.153.11]:60573 "EHLO hosting.visp.net.lb" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751233AbZAKNDa (ORCPT ); Sun, 11 Jan 2009 08:03:30 -0500 In-Reply-To: <20090111125206.GA24173@ioremap.net> Content-Disposition: inline Sender: netdev-owner@vger.kernel.org List-ID: I did test on loaded squid. 49878 connections established ,1421/sec It is not peak time yet, passed 24 hours testing. I can't compare load, because it is real life load and always vary, but thing i can say, it doesn't crash :-) On Sunday 11 January 2009 14:52:06 Evgeniy Polyakov wrote: > On Sat, Jan 10, 2009 at 11:20:19PM -0800, David Miller (davem@davemloft.net) wrote: > > > I'm getting out from the vacations tomorrow and will try to more > > > extensively test the patch (QA group found a bug in ipv6 side, but that > > > was with the .24 backported patch, current one does not have that > > > codepath iirc). But I would like to know if this worth it. > > > > I'll look at it at some point but it won't make this merge window, > > sorry. > > Ok, please drop me a not and if things are good, merge the patch > into the appropriate tree for the next window when time permits.