From owner-freebsd-current  Sun Mar 31 04:42:59 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id EAA23485
          for current-outgoing; Sun, 31 Mar 1996 04:42:59 -0800 (PST)
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 SMTP id EAA23479
          Sun, 31 Mar 1996 04:42:54 -0800 (PST)
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.6.12/8.6.9) id EAA01469; Sun, 31 Mar 1996 04:42:52 -0800
Date: Sun, 31 Mar 1996 04:42:52 -0800
Message-Id: <199603311242.EAA01469@silvia.HIP.Berkeley.EDU>
To: gibbs@freefall.freebsd.org
CC: current@freefall.freebsd.org
In-reply-to: <199603311057.CAA00635@silvia.HIP.Berkeley.EDU> (asami@cs.berkeley.edu)
Subject: Re: aic7xxx driver and parity error problems
From: asami@cs.berkeley.edu (Satoshi Asami)
Sender: owner-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

 * Also, the "disable parity checking" workaround failed here, as I got
 * "timed out in message out phase, SCSISIGI=0xe6" and stuff (I'm not
 * sure, I had to reset the machine while it was still scrolling).
 * 
 * This threw the SCSI system into a serious funk as the next three
 * reboots, from cold start, failed with a solid hang during the probe
 * from ahc0 (I've told you about this before).
 * 
 * It has come up now (with parity checking disabled), so I guess it (the
 * workaround) does work sometimes.

More update.  The machine crashed shortly afterwards (running an old
kernel), and it seemed to have caused something to go seriously wrong.
I haven't been able to boot -current in about 15 tries, using new
kernels, old kernels (unfortunately I have only one version, which is
about 3 days old, which WAS running flawlessly until tonight), parity
turned on/off, etc.  It always dies with the "sd1: timed out in
message in phase, SCSISIGI=0xe6" (note it's now sd1).

However, I managed to boot the -stable system I haven't touched for
months.  The funny thing is, this system is on sd1.  I'm quite
puzzled.

Satoshi