NetBSD Problem Report #3449
Received: (qmail-queue invoked from smtpd); 6 Apr 1997 06:24:29 -0000
Message-Id: <199704060624.BAA03208@hrothgar.gw.com>
Date: Sun, 6 Apr 1997 01:24:26 -0500 (EST)
From: Kimmo Suominen <kim@tac.nyc.ny.us>
Reply-To: kim@netbsd.org
To: gnats-bugs@gnats.netbsd.org
Subject: Kernel detects ahc controllers in wrong sequence
X-Send-Pr-Version: 3.95
>Number: 3449
>Category: port-i386
>Synopsis: Kernel detects ahc controllers in wrong sequence
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: port-i386-maintainer
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Sat Apr 05 22:35:01 +0000 1997
>Closed-Date: Tue Jul 23 10:01:07 +0000 2002
>Last-Modified: Tue Jul 23 10:01:07 +0000 2002
>Originator: Kimmo Suominen
>Release: 19970311
>Organization:
<A HREF="http://www.tac.nyc.ny.us/~kim/">Kimmo Suominen</A>
>Environment:
System: NetBSD hrothgar.gw.com 1.2C NetBSD 1.2C (BEOWULF) #0: Tue Mar 11 20:40:56 EST 1997 kim@ramoth.nyc.deshaw.com:/usr/src/sys/arch/i386/compile/BEOWULF i386
Dell Dimension XPS P166c
Two Adaptec AHA-2940 SCSI controllers
Unfortunately the ahc driver panics, so I had to undo this config,
and cannot reproduce boot messages.
>Description:
The probe order in the kernel is reversed to the BIOS sequence.
I.e. the first controller for BIOS ends up as ahc1 and the second
controller ends up as ahc0.
The problem with this is that C: (i.e. the first drive) does not
end up being sd0 (it became sd1 in my case, but that depends on
how many drives you have on the first controller). Thus booting
is very confused...
Even if the kernel would realize that we booted off of sd3 (or
whatever) it would probably be best to preserve the drive and
controller order from BIOS.
>How-To-Repeat:
Install two ahc controllers and reboot.
>Fix:
I don't have a fix, sorry.
The workaround is to diable the BIOS on the first controller, so
only the drives on the second controller are seen through BIOS.
This gives me C: on the first drive of the second controller, and
this matches what the kernel thinks.
>Release-Note:
>Audit-Trail:
Responsible-Changed-From-To: gnats-admin->port-i386-maintainer
Responsible-Changed-By: fair
Responsible-Changed-When: Mon Dec 28 09:51:41 PST 1998
Responsible-Changed-Why:
This PR is the responsibility of the portmaster,
not the GNATS database administrator.
State-Changed-From-To: open->feedback
State-Changed-By: fair
State-Changed-When: Thu Jan 17 12:32:12 PST 2002
State-Changed-Why:
NetBSD has been through two major releases since this PR was filed;
does the problem still occurr?
State-Changed-From-To: feedback->closed
State-Changed-By: fair
State-Changed-When: Tue Jul 23 02:58:03 PDT 2002
State-Changed-Why:
absent feedback for six months, we declare this PR closed.
It should be noted that it is possible to force controller and drive
probe order with a kernel config that declares the appropriate
values explicitly rather than wildcarding them as GENERIC does, and this
is highly recommended for configurations that use CCD or RAIDFRAME to
guarantee that drives don't "move around" from boot to boot just because
one drive on a SCSI bus was off or not responding at boot time.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.