Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 26 Apr 2006 00:45:55 +0900 (JST)
From:      Munehiro Matsuda <haro@h4.dion.ne.jp>
To:        daichi@freebsd.org
Cc:        ozawa@ongs.co.jp, freebsd-hackers@freebsd.org, freebsd-fs@freebsd.org, freebsd-current@freebsd.org, Alexander@Leidinger.net, saturnero@freesbie.org
Subject:   Re: [ANN] unionfs patchset-11 release
Message-ID:  <20060426.004555.21603794.haro@h4.dion.ne.jp>
In-Reply-To: <444E13BA.8050902@freebsd.org>
References:  <4417DD8D.3050201@freebsd.org> <4433CA53.5050000@freebsd.org> <444E13BA.8050902@freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi Goto-san,

From: Daichi GOTO <daichi-at-freebsd.org>
Date: Tue, 25 Apr 2006 21:19:06 +0900
::It is my pleasure and honor to announce the availability of
::the unionfs patchset-11.
<snip>
::   Heads Up: Above English text contains some Japanese text.
::             If you are interested in translating the Japanese
::             text into English, please do it and send it to me.

How about this?

=--------------------------------------------------------------------------
On creating a hardlink, due to VOP_GETWRITEMOUNT() called right after
lookup(), if only files exist in the underlying filesystem,
VOP_GETWRITEMOUNT() cannot be process on the vnode for the layer above.
Thus creating hardlink fails.
A complete fix is to change VOP_GETWRITEMOUNT() to create a shadowfile,
but its not possible due to the fact that, locking status of vnode
passed as argument to VOP_GETWRITEMOUNT() is not stable.
In the current kernel code base, filesystems using vop_stdgetwritemount()
simply aquire mount point vnode, thus using vnode for directory above
seems to suffice and used for the current implementation.
But this logic cannot be considered safe and may cause problems in the
future. Also, when mounting a directory containing multiple mount points
using unionfs, becomes evil. Thus, it is dangerous to mount directories
containing multiple mount points, using the current implementation of
unionfs.
=--------------------------------------------------------------------------

Hope this helps,
 Haro
=------------------------------------------------------------------------------
           _ _    Munehiro (haro) Matsuda
 -|- /_\  |_|_|   Internet Solution Dept., KGT Inc.
 /|\ |_|  |_|_|   2-8-8 Shinjuku Shinjuku-ku Tokyo 160-0022, Japan
                  Tel: +81-3-3225-0767  Fax: +81-3-3225-0740
                  Email: haro-at-kgt.co.jp



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060426.004555.21603794.haro>