From owner-freebsd-bugs  Tue Apr 30 01:42:35 1996
Return-Path: owner-bugs
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id BAA19254
          for bugs-outgoing; Tue, 30 Apr 1996 01:42:35 -0700 (PDT)
Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id BAA19249
          for <bugs@freebsd.org>; Tue, 30 Apr 1996 01:42:31 -0700 (PDT)
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.5/8.6.9) id BAA06847; Tue, 30 Apr 1996 01:41:46 -0700 (PDT)
Date: Tue, 30 Apr 1996 01:41:46 -0700 (PDT)
Message-Id: <199604300841.BAA06847@silvia.HIP.Berkeley.EDU>
To: bridges@pioneer.arc.nasa.gov
CC: bugs@freebsd.org, bridges@pioneer.arc.nasa.gov
In-reply-to: <199604300547.WAA27896@pioneer.arc.nasa.gov> (bridges@pioneer.arc.nasa.gov)
Subject: Re: Prob - 2.1 FreeBSD boot floppy "can't see" controller: Adaptec 2940W
From: asami@cs.berkeley.edu (Satoshi Asami)
Sender: owner-bugs@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

 * This pentium system has an Adaptec 2940W SCSI controller with 
 * Seagate 32550W 2GB Barrcuda disk on an Intel Endeavor motherboard.
 * According the 2.1 FreeBSD docs, this controller is supported.

Yes.  I've installed some 2.1R systems over here using a 2940UW with
wide or narrow disks.  We never had problems with the adapter probe.

 * Based on the FreeBSD documentation I should have seen "ahc0" come up 
 * when devices are being probed at boot time.  This didn't happen.  
 * I tried a couple of reset boots, but still no disk controller.

It could be something with the motherboard's setup.  We haven't tried
Intel motherboards.  Does the adaptec's boot BIOS ("SCSISelect") come
up and list all the disks?

 * In any case, I next floppy booted with the "-c" option.
 * The config "ls" command did not show the expected "ahc0". 

This is natural.  PCI devices using proper PCI probe routines don't
show up here, they don't have any I/O address or interrupts
pre-specified.

 * Visual mode shows that device "ahc1" should take care of the 
 * 294x family of controllers  --> unlike the docs which refer to "ahc0".

The ahc1 was in the 2.1R kernel for the EISA (274x) and/or VLB (284x)
version, I believe.  This is not what you need.

Satoshi