From owner-freebsd-stable@FreeBSD.ORG Tue Jun 6 11:38:00 2006 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 74DC416A566 for ; Tue, 6 Jun 2006 11:38:00 +0000 (UTC) (envelope-from petefrench@ticketswitch.com) Received: from mail.ticketswitch.com (mail.ticketswitch.com [194.200.93.188]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0D69543D45 for ; Tue, 6 Jun 2006 11:37:59 +0000 (GMT) (envelope-from petefrench@ticketswitch.com) Received: from [172.16.1.6] (helo=dilbert.firstcallgroup.co.uk) by mail.ticketswitch.com with esmtp (Exim 4.60 (FreeBSD)) (envelope-from ) id 1FnZsk-000GgA-PJ; Tue, 06 Jun 2006 12:37:58 +0100 Received: from petefrench by dilbert.firstcallgroup.co.uk with local (Exim 4.61 (FreeBSD)) (envelope-from ) id 1FnZsk-0007i7-DE; Tue, 06 Jun 2006 12:37:58 +0100 To: jfvogel@gmail.com In-Reply-To: <2a41acea0606051012r5831971qb3d2fb88e701225@mail.gmail.com> Message-Id: From: Pete French Date: Tue, 06 Jun 2006 12:37:58 +0100 Cc: freebsd-stable@freebsd.org Subject: Re: Dual Opteron system will not run SMP 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: Tue, 06 Jun 2006 11:38:00 -0000 > We (Intel networking group) have seen issues with SMP and amd64 > as well, although I dont know that its fully characterized (what works > and what doesnt) yet. When I tried mine under amd64 it seemed to behave slightly better oddly enough - it would at least boot up the SMP kernel without ACPI, though it would not recognise the second processor. The other difference when I was doing that test was that I was using an IDE drive though. I suppose I really need to put amd64 onto a normal disc drive and try booting that to narrow down which change made the difference there. > If just non-ACPI isnt sufficient, the other thing SAFE does is turn > off disk DMA. I have an as-yet unreleased system that has this > same type of issue, and the problem is that two PCI device ID's > are not recognized, so maybe that will be your problem. Interesting - I didn't realise you could do that with SCSI controllers. I had been wondering what 'safe' does differently, and whether I could fill up loader.conf with approrpiate values to emulate what 'safe' does and then start commenting them out one at a time. I'll do some more experiments and report back. I don't particularly mind if it doesn't work in the end (I'll just sell the pair of Opterons on ebay and buy a single faster one), but it's always good to try and track down a bug, so thanks for taking an interest. -pete.