From owner-freebsd-hackers  Sun Aug 31 22:56:33 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id WAA08481
          for hackers-outgoing; Sun, 31 Aug 1997 22:56:33 -0700 (PDT)
Received: from hq.icb.chel.su (hq.icb.chel.su [193.125.10.33])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA08394
          for <hackers@freebsd.org>; Sun, 31 Aug 1997 22:52:09 -0700 (PDT)
Received: (babkin@localhost) by hq.icb.chel.su (8.8.3/8.6.5) id LAA16143 for hackers@freebsd.org; Mon, 1 Sep 1997 11:35:14 +0600 (ESD)
From: "Serge A. Babkin" <babkin@hq.icb.chel.su>
Message-Id: <199709010535.LAA16143@hq.icb.chel.su>
Subject: NPX old bug
To: hackers@freebsd.org
Date: Mon, 1 Sep 1997 11:35:13 +0600 (ESD)
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Sender: owner-freebsd-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Hi!

I've discovered an interesting bug in `npx' driver in an old
version (2.2-SNAP of spring 1996). When it is the last device
that adds its kdc_* information at boot time and then
later some device tries to add it information when the
system is running (for example in result of `scsi -p' command)
it remaps the npx driver code to some empty page.

After I found this reason of panic I checked -current and
found that npx just has no kdc_* structure anymore ! Did
someone else found these panics before me or is it
just a coincidence ? In the last case I think the bug
may reappear in some other driver.

-SB

P.S. I've found that `vtophys' is worth adding it to DDB. The
addition is very simple. Does anyone from the core team think
that it's worth too ?