From owner-freebsd-current  Fri Nov 29 21:37:37 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id VAA24917
          for current-outgoing; Fri, 29 Nov 1996 21:37:37 -0800 (PST)
Received: from sovcom.kiae.su (sovcom.kiae.su [193.125.152.1])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id VAA24912;
          Fri, 29 Nov 1996 21:37:34 -0800 (PST)
Received: by sovcom.kiae.su id AA16660
  (5.65.kiae-1 ); Sat, 30 Nov 1996 08:25:18 +0300
Received: by sovcom.KIAE.su (UUMAIL/2.0); Sat, 30 Nov 96 08:25:18 +0300
Received: (from ache@localhost) by nagual.ru (8.8.3/8.8.3) id IAA00552; Sat, 30 Nov 1996 08:24:55 +0300 (MSK)
Message-Id: <199611300524.IAA00552@nagual.ru>
Subject: Re: MSDOS lkm just broken (MAXBSIZE)
In-Reply-To: <199611300214.VAA09459@dyson.iquest.net> from "John S. Dyson" at "Nov 29, 96 09:14:02 pm"
To: toor@dyson.iquest.net (John S. Dyson)
Date: Sat, 30 Nov 1996 08:24:55 +0300 (MSK)
Cc: toor@dyson.iquest.net, current@freebsd.org, dyson@freebsd.org
From: =?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= (Andrey A. Chernov) <ache@nagual.ru>
Organization: self
X-Class: Fast
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> THE SYSTEM NEVER SUPPORTED 32K out of the BOX!!!  MSDOS WAS ASKING
> FOR THE IMPOSSIBLE!!!

MSDOS asking for impossible because you instruct it to do so:

msdosfs_vfsops.c:		pmp->pm_fatblocksize = MAXBSIZE;

MAXBZSIZE = 32K for MSDOS lkm, because MSDOSFS is defined


> Sorry, but unless you were running with MAXBSIZE=32K you were
> heading for trouble.  You can remove the check(panic), and get exactly
> what you would have before...  A potentially corrupted filesystem.  The
> only reason that you have not had a corrupted file system is by
> chance...  Sorry again, that is the situation.  I am working on
> a real solution, and hackery is NOT in the plan.

It seems that

1) you introduce this corruption first
2) then check for it and panic

For old variant

msdosfs_vfsops.c:		pmp->pm_fatblocksize = MAXBSIZE;

was evaluated to 16K and not cause corruption.

-- 
Andrey A. Chernov
<ache@nagual.ru>
http://www.nagual.ru/~ache/