From owner-freebsd-amd64@FreeBSD.ORG Sun Feb 1 16:21:20 2004 Return-Path: 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 E2DFA16A4CE for ; Sun, 1 Feb 2004 16:21:20 -0800 (PST) Received: from omega.metrics.com (ns.metrics.com [204.138.110.200]) by mx1.FreeBSD.org (Postfix) with ESMTP id 958E943D5A for ; Sun, 1 Feb 2004 16:21:06 -0800 (PST) (envelope-from tomh@waterloo.equitrac.com) Received: from syncro.metrics.com ([192.168.96.20]) by omega.metrics.com (8.9.3/8.9.3) with ESMTP id TAA08278 for ; Sun, 1 Feb 2004 19:21:06 -0500 (EST) Received: by SYNCRO with Internet Mail Service (5.5.2653.19) id ; Sun, 1 Feb 2004 19:19:39 -0500 Message-ID: From: "Haapanen, Tom" To: amd64@freebsd.org Date: Sun, 1 Feb 2004 19:19:34 -0500 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: text/plain Subject: RE: Dual processor, AMD 64 machine freezing. X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 02 Feb 2004 00:21:21 -0000 Whoops ... looks like I got my aac and asr mixed up! Sorry about that ... As for the memory allocation ... is that something that FreeBSD *could* do in the future? Would that fit in with the kernel architecture? And if that's not the case today, is there any advantage to balancing the memory between the two CPUs? Tom -----Original Message----- From: Andreas Braukmann [mailto:braukmann@tse-online.de] Sent: Sunday 01 February 2004 18:52 To: Haapanen, Tom Cc: amd64@freebsd.org Subject: RE: Dual processor, AMD 64 machine freezing. On 02/01/04 18:34:10 -0500 Haapanen, Tom wrote: > I run a K8S Pro with dual 244s (couldn't find any 246s and the 248s > were just too dear for me) and 2 GB ... and I started off with an > Adaptec 2100S RAID, but had consistent panics when booting. > Apparently the aac driver isn't exactly 64-bit clean. The 2100S isn't driven by aac(4) but by asr(4). The asr-driver is in fact not 64-bit clean (as Scott Long stated multiple times :-/). The aac-driver is perfectly fine with the 64 bits (for me). It survives heavy i/o load without a hitch. I ran quite a bunch of tests (parallel bonnie++ spreaded over multiple raid volumes and multiple file systems, postmark, etc.) with my 2200S, 2120S and 5400S. > P.S. For best performance, I think you really want to run 4x 512 MB. > Running with two DIMMs means either you only get 64-bit memory access > (not > 128-bit) or else you need to put both DIMMs into the CPU1 memory slots > (which means CPU2 will have to access those through hypertransport). Thats theory. ;-) Since the allocators don't know about the numa-like architecture memory would be accessed through hyptertransport (statistically) more or less "half of the time". (CPU0 ---> HT ---> MEM1 ; CPU1 ---> HT ---> MEM0) -Andreas