From owner-freebsd-stable@FreeBSD.ORG Wed Feb 27 20:16:24 2008 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6F9551065673 for ; Wed, 27 Feb 2008 20:16:24 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost1.sentex.ca (smarthost1.sentex.ca [64.7.153.18]) by mx1.freebsd.org (Postfix) with ESMTP id 4120E8FC1C for ; Wed, 27 Feb 2008 20:16:24 +0000 (UTC) (envelope-from mike@sentex.net) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost1.sentex.ca (8.14.2/8.14.2) with ESMTP id m1RKG2Op079855; Wed, 27 Feb 2008 15:16:02 -0500 (EST) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.13.8/8.13.3) with ESMTP id m1RKG1tl061076 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 27 Feb 2008 15:16:01 -0500 (EST) (envelope-from mike@sentex.net) Message-Id: <200802272016.m1RKG1tl061076@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Wed, 27 Feb 2008 15:13:59 -0500 To: Holger Kipp , freebsd-stable@freebsd.org From: Mike Tancsa In-Reply-To: <20080227200923.GA28690@intserv.int1.b.intern> References: <20080227104942.GA14530@intserv.int1.b.intern> <200802271452.m1REqJ6H059687@lava.sentex.ca> <20080227195058.GA27997@intserv.int1.b.intern> <20080227200923.GA28690@intserv.int1.b.intern> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Cc: Subject: Re: [solved!] Re: em very slow, shared irq... on 6.2p8 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Feb 2008 20:16:24 -0000 At 03:09 PM 2/27/2008, Holger Kipp wrote: >On Wed, Feb 27, 2008 at 08:50:58PM +0100, Holger Kipp wrote: > > On Wed, Feb 27, 2008 at 09:50:16AM -0500, Mike Tancsa wrote: > > > > more details below. as it currently is, polling seems to do > > the trick, however handling several em-interfaces with the > > same irq (mind you, it is pci) shouldn't cause delays of > > up to 1.5 seconds for a simple ping... Therefore I consider > > using polling for a nearly idle system more a workaround > > than a solution to this problem :-( >[...] > > with msi enabled (via sysctl) I get > > > > 33 packets transmitted, 33 packets received, 0% packet loss > > round-trip min/avg/max/stddev = 1.865/156.421/1339.841/239.375 ms > > > > so looks equally bad (I don't consider 30-40 packets a meaningful sample). > > I don't know if it makes any differences if switched on directly in > > loader.conf, though. > >have now activated msi in loader.conf and get very good results again. > >38 packets transmitted, 38 packets received, 0% packet loss >round-trip min/avg/max/stddev = 0.291/0.425/0.595/0.067 ms > >without polling activated. So it was msi after all I needed here. >Maybe this should go into docu for em or ifconfig? Hi, Yes, sorry I should have mentioned, you need to reboot. But I strongly suggest upgrading to 6.3R as there are a number of em bugs that are fixed.... Perhaps some IRQ issues as well. But for MSI in general, I think the Intel guy recommended running that way for the NIC. ---Mike