From owner-freebsd-amd64@FreeBSD.ORG Fri Jul 29 19:10:56 2005 Return-Path: X-Original-To: freebsd-amd64@freebsd.org Delivered-To: freebsd-amd64@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C822016A420 for ; Fri, 29 Jul 2005 19:10:56 +0000 (GMT) (envelope-from vivek@khera.org) Received: from yertle.kcilink.com (yertle.kcilink.com [65.205.34.180]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9990F43D4C for ; Fri, 29 Jul 2005 19:10:53 +0000 (GMT) (envelope-from vivek@khera.org) Received: from [192.168.7.103] (host-103.int.kcilink.com [192.168.7.103]) by yertle.kcilink.com (Postfix) with ESMTP id 8A049B869 for ; Fri, 29 Jul 2005 15:10:52 -0400 (EDT) Mime-Version: 1.0 (Apple Message framework v733) In-Reply-To: <1122594648.45827.11.camel@server.mcneil.com> References: <3.0.1.32.20050728013152.00a4d188@pop.redshift.com> <20050728193519.A66123@cons.org> <1122594648.45827.11.camel@server.mcneil.com> Content-Type: multipart/signed; micalg=sha1; boundary=Apple-Mail-8-475646239; protocol="application/pkcs7-signature" Message-Id: From: Vivek Khera Date: Fri, 29 Jul 2005 15:10:50 -0400 To: freebsd-amd64 List X-Mailer: Apple Mail (2.733) X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Subject: Re: Benchmarks: AMD64 vs i386 on Dual 246 Opteron X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Jul 2005 19:10:56 -0000 --Apple-Mail-8-475646239 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed On Jul 28, 2005, at 7:50 PM, Sean McNeil wrote: > On Thu, 2005-07-28 at 19:35 -0400, Martin Cracauer wrote: > >> I highly suspect that the 64 bit version of FreeBSD in this >> comparision was somehow doomed before it started. Opportunities are >> leaving in kernel safety code in one of the kernels, a driver that >> just doesn't work right on AMD64. Or the same for any of the >> software, maybe php is hosed in its 64 bit build. >> > > If network throughput can effect the testing, I would suspect network > drivers. I've found the re0 driver performs horrible on amd64. > Especially with gigE. The OP used a S2881 mobo, which has a bge ethernet. I've had nothing but bad luck with that in FreeBSD/amd64. I put in some intel NICs and everything got better and more stable. Vivek Khera, Ph.D. +1-301-869-4449 x806 --Apple-Mail-8-475646239--