From owner-freebsd-current  Tue Jun 24 10:00:52 1997
Return-Path: <owner-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id KAA25164
          for current-outgoing; Tue, 24 Jun 1997 10:00:52 -0700 (PDT)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA25157
          for <current@FreeBSD.ORG>; Tue, 24 Jun 1997 10:00:47 -0700 (PDT)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id CAA28490; Wed, 25 Jun 1997 02:59:19 +1000
Date: Wed, 25 Jun 1997 02:59:19 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199706241659.CAA28490@godzilla.zeta.org.au>
To: bde@zeta.org.au, kato@migmatite.eps.nagoya-u.ac.jp
Subject: Re: vm_bounce_alloc: Unmapped page
Cc: current@FreeBSD.ORG
Sender: owner-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>> ext2fs is the only other file system that supports clustered writes,
>> and its doclusterwrite variable is hidden in an ifdef tangle.
>
>Why doclusterread and doclusterwrite in ext2fs are macroized?  They
>cannot be changed with sysctl :-(.

ext2fs is not up to date.

>Does ext2fs have same problem as ufs for accesing unionfs?  I cannot
>test ext2fs by myself becasue I have no Linux partition.

I would expect it to have exactly the same problem unless the problem
is related to block sizes or fragments.  ext2fs is very similar to ufs
except for block allocation and mapping.

Bruce