From owner-aic7xxx@FreeBSD.ORG Thu Nov 6 12:06:32 2003 Return-Path: Delivered-To: aic7xxx@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E162416A4CE for ; Thu, 6 Nov 2003 12:06:32 -0800 (PST) Received: from proxy1.castlebranch.com (proxy1.castlebranch.com [66.21.56.81]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6F08C43FEC for ; Thu, 6 Nov 2003 12:06:31 -0800 (PST) (envelope-from wendels@castlebranch.com) Received: from penguin6.castlebranch.com (penguin6.castlebranch.com [192.168.2.29]) by proxy1.castlebranch.com (8.11.6/8.11.6) with ESMTP id hA6K6Un17049 for ; Thu, 6 Nov 2003 15:06:30 -0500 From: Wendell Smith To: aic7xxx@freebsd.org Content-Type: text/plain Organization: CastleBranch, Inc Message-Id: <1068149172.25505.238.camel@penguin6> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.3 Date: 06 Nov 2003 15:06:15 -0500 Content-Transfer-Encoding: 7bit Subject: Question About Bus Speeds (and SCSI in general) X-BeenThere: aic7xxx@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Adaptec Device Drivers in FreeBSD and Linux List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 06 Nov 2003 20:06:33 -0000 I am trying to tune a certain server and am having trouble getting my head wrapped around the fundamentals of it's scsi disk/bus system. The server in question is a Netfinity 7000 M10 with a ServeRAID 3H Raid controller. Let me try to provide a little background to better place my confusion in your heads... OS: RedHat 9.0 Kernel: 2.4.20-20.9smp dmesg output: scsi0 : IBM PCI ServeRAID 5.10.21 Vendor: IBM Model: SERVERAID Rev: 1.00 Type: Direct-Access ANSI SCSI revision: 02 Vendor: IBM Model: SERVERAID Rev: 1.00 Type: Direct-Access ANSI SCSI revision: 02 Vendor: IBM Model: SERVERAID Rev: 1.00 Type: Processor ANSI SCSI revision: 02 Vendor: IBM Model: SAF-TE Rev: 2 Type: Processor ANSI SCSI revision: 02 Attached scsi disk sda at scsi0, channel 0, id 0, lun 0 Attached scsi disk sdb at scsi0, channel 0, id 1, lun 0 SCSI device sda: 17772544 512-byte hdwr sectors (9100 MB) Partition check: sda: sda1 sda2 sda3 sda4 < sda5 sda6 > SCSI device sdb: 71686144 512-byte hdwr sectors (36703 MB) sdb: sdb1 sdb2 scsi1 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.8 aic7895C: Ultra Wide Channel B, SCSI Id=7, 32/253 SCBs scsi2 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.8 aic7895C: Ultra Wide Channel A, SCSI Id=7, 32/253 SCBs Is this saying that I have three SCSI busses? I initially looked at /proc/scsi/aic7xxx/1 and /proc/scsi/aic7xxx/2 and found that the bus speed for each had been set to 40 MB/sec. This set off a flag in my head because I had set the ServerRAID controller to Ultra 2 SCSI speed which should be 80 MB/sec. Right? So if I have three SCSI busses, are they fully independent of each other with respect to the bus speed settings? How does the aic7xxx bus speed get set? Does it matter to me in this case? Would I be better off utilizing the 2nd and 3rd SCSI bus with respect to this server and it's disks? (Both disks are set to Raid level 0) The reason behind this line of questioning is that when I execute: hdparm -t /dev/sd[a-b] I get a results of around 28 MB/sec I would expect more from a SCSI system set to run at Ultra 2 speeds. By the way, both disks are Ultra 2 rated so I don't feel there is an issue there. Any help with these question would be greatly appreciated. Any documentation that you can point me to (with respect to SCSI system within the context of Linux) would also be greatly appreciated. Regards, Wendell Smith