From owner-freebsd-scsi Wed Oct 28 23:28:55 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA14593 for freebsd-scsi-outgoing; Wed, 28 Oct 1998 23:28:55 -0800 (PST) (envelope-from owner-freebsd-scsi@FreeBSD.ORG) Received: from dante.visionaire.net (visionaire.ping.de [195.37.123.61]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA14586; Wed, 28 Oct 1998 23:28:44 -0800 (PST) (envelope-from thomas@visionaire.ping.de) Received: from thomas by dante.visionaire.net with local (Exim 1.92 #1 (Debian)) id 0zYn0e-0000KU-00; Thu, 29 Oct 1998 08:00:44 +0000 Message-ID: <19981029080043.C812@visionaire.ping.de> Date: Thu, 29 Oct 1998 08:00:43 +0000 From: Thomas Keusch To: Howard Lew , freebsd-scsi@FreeBSD.ORG, freebsd-questions@FreeBSD.ORG Cc: freebsd-scsi@FreeBSD.ORG, freebsd-questions@FreeBSD.ORG Subject: Re: NCR 53c875 SCSI Problems References: <19980928053655.19907@visionaire.ping.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.91.2 In-Reply-To: ; from Howard Lew on Wed, Sep 30, 1998 at 11:40:42PM -0700 Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, Sep 30, 1998 at 11:40:42PM -0700, Howard Lew wrote: > > I have a very similar/the same problem with a SymbiosLogic > > 8750SP Ultra-SCSI controller and two IBM DCAS 4.1 Gb HDDs. > > > > I did not find a solution yet, but it seems that the problem on > > my box is related to FreeBSD trying to find out the size of the disks. > > This produces basically the same error as yours, but quits scanning > > with an "could not get size" error for each disk after a while. > > Yes, this is exactly the same problem I am having. I guess if I wait a > very long time, it might get past the boot probe after failing to detect > the affected drives, but generally the boot probe shouldn't take 1 hour or > more to do. Did you ever try if it gets past that state? If so, how long did it take? > > I took of the entire bus from the controller, so that the controller > > was the only SCSI device in the system. FreeBSD doesn't choke on the > > controller itself, it's the disks. > > Yes, it is the disks -- not the controller. > > There is an easy fix if you already have a working FreeBSD system that > is booting off another drive. As long as you are not using the WIDE bus, > you can force everything to 8 bit mode. I found that changing the > SCSI_NCR_MAX_WIDE to 0 does the trick. Of course, this forces 8 bit > mode and disables the WIDE bus, so you should not have any devices on the > WIDE bus. I don't have ANY Wide devices. The controller is Ultra SCSI only, but has a Wide capable chipset nevertheless. So this fix will work for me, the only drawback being that I have to install do an IDE disk first, learning how to compile the kernel first, ... etc. > As it is, this is not a real fix because you need a working system to > make the changes in ncr.c and then recompile the system. So if a new > user wants to install FreeBSD on his hard disk, this quick fix will not > work as they won't be able to get through the installation. I've been using Linux for a year now, so I guess I have the experience to get my system working, but I'm still dependant on an IDE disk, which I luckily still have. A totally inexperienced new user would have at least a really hard time to install onto SCSI disks under these circumstances, that's right. Odds are, that there are more narrow SCSI controllers out there which have this "feature" of a Wide chipset, so I guess an automatic fallback to narrow mode would be a good thing(tm). Anyway, I will try your fix. Thank you for that. Have a nice day -- thomas. .powered.by.debian/linux. irc.:.#meeting.points, #frust.ger To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message