From owner-freebsd-net@FreeBSD.ORG Tue Nov 20 23:30:23 2012 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id F2B43B52 for ; Tue, 20 Nov 2012 23:30:22 +0000 (UTC) (envelope-from barney_cordoba@yahoo.com) Received: from nm15.bullet.mail.ne1.yahoo.com (nm15.bullet.mail.ne1.yahoo.com [98.138.90.78]) by mx1.freebsd.org (Postfix) with ESMTP id 9E8568FC13 for ; Tue, 20 Nov 2012 23:30:22 +0000 (UTC) Received: from [98.138.90.50] by nm15.bullet.mail.ne1.yahoo.com with NNFMP; 20 Nov 2012 23:30:15 -0000 Received: from [98.138.89.163] by tm3.bullet.mail.ne1.yahoo.com with NNFMP; 20 Nov 2012 23:30:15 -0000 Received: from [127.0.0.1] by omp1019.mail.ne1.yahoo.com with NNFMP; 20 Nov 2012 23:30:15 -0000 X-Yahoo-Newman-Property: ymail-3 X-Yahoo-Newman-Id: 919819.81003.bm@omp1019.mail.ne1.yahoo.com Received: (qmail 21516 invoked by uid 60001); 20 Nov 2012 23:30:15 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1353454215; bh=UFN+Q2Pm+HvYxArpSLyvh9kCbu79RpS2E3JHZ0oE5hM=; h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:Message-ID:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=NDsm+cK1gRnjAHRIAqMu3tM5C6K2XQB8fszMHuXVSTKTLbebW+vntfosdxasV6qcqcLPwRsR8LJsHZr9egrz2KADms52WP403ZGEaRD9ijw2/5zvZZRuVoCIj8tI75CHU2B3v/bFW09WKmasyMU18zsGxQ9fsNXNwLBncciEzcA= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:Message-ID:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=Y4FY2qbDxxP8Gw+x0o70A7Bp7/1rTyLQx2Ol29F1d5z/pFqjGULnke+WQEqO3lapkTcm3sJu+kKPR2D3knyC+L7I251lXORhJy3USSkpEguPXZoV4/MT+Df39zMCnglqJnQBYkMosoCjjzOh78Py7SvbRVd9Fi/DXIJvDo1J2a8=; X-YMail-OSG: W72zAu4VM1lFcoyTs0fLc9fnlfGzNOp1hSHwPt6JZHNlaKy YP2qZbvB.MIU7bPgdtCxfPb3iHmWetde7czaRV.gDEgPkbPgLGOYqxAbfR.G ITvP9jVdPgbez2aLRed6KzziVftWU_AwRt.FAJMgDp8TtndAhBvLO619Wrk9 8LR7L.1Fu5xes.PKp2V4z3cfvqknks7nyeUSAcdAoND4if0MI9aJbLahUjnF k3nMglqK.Iakbk.95EkWqUi9CjIVAUyf4YX2GBy5FIXfFLcxTfr1FLmG03rO 9MpotBd0RRlQysDaG1kxG9y8eN7qqFn7VEoaHsdDoUwA76hpAgO0fc_jKtGZ A7Y1E7.GjHO12nIfBSrga.D6HsQ.8dNAhaTptBmcYfmnRtv0FUaMZ0UaJEWT dHQEUMozmMW_M5465N2rkqJPYCuoBWTHkavYu7tRIXPX3a65VrzPgAQy7tdp aUXKo4_R.Z.07pKVQUm3uBgp3mTPinQmyIIHunSuyrB8AG3rhEpJl_AJmEaf WQNmKBnHEziGZhC_D2Xsr..0ct0d4ag-- Received: from [174.48.128.27] by web121601.mail.ne1.yahoo.com via HTTP; Tue, 20 Nov 2012 15:30:15 PST X-Rocket-MIMEInfo: 001.001, CgotLS0gT24gVHVlLCAxMS8yMC8xMiwgSW5nbyBGbGFzY2hiZXJnZXIgPGlmQHhpcC5hdD4gd3JvdGU6Cgo.IEZyb206IEluZ28gRmxhc2NoYmVyZ2VyIDxpZkB4aXAuYXQ.Cj4gU3ViamVjdDogUmU6IEZyZWVCU0QgYm94ZXMgYXMgYSAncm91dGVyJy4uLgo.IFRvOiBmcmVlYnNkLW5ldEBmcmVlYnNkLm9yZwo.IERhdGU6IFR1ZXNkYXksIE5vdmVtYmVyIDIwLCAyMDEyLCA2OjA0IFBNCj4gQW0gMjAuMTEuMjAxMiAyMzo0OSwgc2NocmllYiBBbGZyZWQKPiBQZXJsc3RlaW46Cj4gPiBPbiAxMS8yMC8xMiAyOjQBMAEBAQE- X-Mailer: YahooMailClassic/15.0.8 YahooMailWebService/0.8.126.470 Message-ID: <1353454215.20382.YahooMailClassic@web121601.mail.ne1.yahoo.com> Date: Tue, 20 Nov 2012 15:30:15 -0800 (PST) From: Barney Cordoba Subject: Re: FreeBSD boxes as a 'router'... To: Ingo Flaschberger In-Reply-To: <50AC0C92.8080603@xip.at> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Cc: freebsd-net@freebsd.org X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 20 Nov 2012 23:30:23 -0000 =0A=0A--- On Tue, 11/20/12, Ingo Flaschberger wrote:=0A=0A> Fro= m: Ingo Flaschberger =0A> Subject: Re: FreeBSD boxes as a 'route= r'...=0A> To: freebsd-net@freebsd.org=0A> Date: Tuesday, November 20, 2012,= 6:04 PM=0A> Am 20.11.2012 23:49, schrieb Alfred=0A> Perlstein:=0A> > On 11= /20/12 2:42 PM, Jim Thompson wrote:=0A> >> On Nov 20, 2012, at 3:52 PM, Bar= ney Cordoba =0A> wrote:=0A> >> =0A> >> You're ent= itled to your opinion, but experimental=0A> results have tended to show you= rs incorrect.=0A> >> =0A> >> Jim=0A> > Agree with Jim.=A0 If you want pure = packet=0A> performance you burn a core to run a polling loop. =0A> =0A> At = new systems, without polling I had better performance and=0A> no live-locks= ,=0A> at old systems (Intel 82541GI) polling prevent live-locks.=0A> =0A> B= est test:=0A> Loop a GigE Switch, inject a Packet and plug it into the=0A> = test-box.=0A=0AYeah, thats a good real-world test.=0A=0ATo me "performance"= is not "burning a cpu" to get some extra pps. =0APerformance is not droppi= ng buckets of packets. Performance is using=0Aless cpu to do the same amoun= t of work. =0A=0AIs a machine that benchmarks at 998Mb/s at 95% cpu really = a "higher=0Aperformance" system than one that does 970Mb/s and uses 50% of = the cpu?=0A=0AThe measure of performance is to manage an entire load withou= t dropping=0Aany packets. If your machine goes into live-lock, then you nee= d more=0Amachine. Hacking it so that it drops packets is hardly a solution.= =0A=0ABC