From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 25 May 2001 06:00:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 25 May 2001 06:00:12 -0400 Received: from pc7.prs.nunet.net ([199.249.167.77]:62734 "HELO pc7.prs.nunet.net") by vger.kernel.org with SMTP id ; Fri, 25 May 2001 05:59:53 -0400 Message-ID: <20010525095952.15078.qmail@pc7.prs.nunet.net> From: "Rico Tudor" Subject: Re: Big-ish SCSI disks To: gjohnson@research.canon.com.au (Greg Johnson) Date: Fri, 25 May 101 04:59:52 -0500 (CDT) Cc: linux-kernel@vger.kernel.org In-Reply-To: <20010525062236.8CC1837530@zapff.research.canon.com.au> from "Greg Johnson" at May 25, 1 04:22:36 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I'm running stock 2.4.4 on five PCs with these features: ServerWorks III HE, 2x 933MHz, 4GB RAM, dual-channel sym53c896 (FAST-40 WIDE) SCSI controller. One PC has the new 181GB Seagate SCSI drive; another uses a 3ware RAID controller with 4x 40GB IDE (looks like a 160GB SCSI drive). All is fine with slackware-current. You might play with "nosmp" and "noapic" kernel params. When building a kernel, disable everything including SCSI: work your way up from IDE or floppy root. SMP failed on our systems until I dumped the slocket-based CPUs for Slot 1. (Still a mystery, that.)