From owner-freebsd-hackers Sat Apr 8 09:11:38 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA17759 for hackers-outgoing; Sat, 8 Apr 1995 09:11:38 -0700 Received: from xi.dorm.umd.edu (xi.dorm.umd.edu [129.2.140.12]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id JAA17753 for ; Sat, 8 Apr 1995 09:11:37 -0700 Received: (from smpatel@localhost) by xi.dorm.umd.edu (8.6.11/8.6.9) id MAA12740; Sat, 8 Apr 1995 12:11:22 -0400 Date: Sat, 8 Apr 1995 12:11:22 -0400 (EDT) From: Sujal Patel X-Sender: smpatel@xi.dorm.umd.edu To: "Rodney W. Grimes" cc: Wilko Bulte , hackers@FreeBSD.org Subject: Re: large filesystems/multiple disks [RAID] In-Reply-To: <199504081554.IAA15309@gndrsh.aac.dev.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hackers-owner@FreeBSD.org Precedence: bulk On Sat, 8 Apr 1995, Rodney W. Grimes wrote: > No, they just don't perform well. The scsi command over head is high > on them. I am not sure if this applies to the 24F, but the 34F is a > bus pig (it grabbs the vlb bus for long periods of time, locking out > other interrupts :-(). That is very true, the card is a hog. My ISA busmaster ethernet card spews errors all the time during heavy disk activity. The U24F has an option in the EISA config which can help this (Maximum Bus Hold Time? or something like that)... I think it's set to 500us(ns?) by default, You can lower it to 250us(ns) which helps some. Do you think that there is anything that can be done in the driver that could help this? Also, I know that U24F Bios Version 2.01+ has a lot of fixes in the Scatter/Gather Code on the card. Can someone with a card with 2.01+ Bios give us some benchmarks please? Sujal