Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 28 Nov 1995 14:30:12 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        bde@zeta.org.au (Bruce Evans)
Cc:        chuck@fang.cs.sunyit.edu, hackers@freebsd.org
Subject:   Re: MsDoss FileSystem
Message-ID:  <199511282130.OAA22102@phaeton.artisoft.com>
In-Reply-To: <199511281505.CAA01593@godzilla.zeta.org.au> from "Bruce Evans" at Nov 29, 95 02:05:01 am

next in thread | previous in thread | raw e-mail | index | archive | help
> >support? Does BSD/OS, NetBSD, Lites suffer from the same problems? 
> 
> NetBSD has kept it up to date better.  E.g., in 4.4lite introduced waiting
> on VXLOCK in ufs_lock().  This change has propagated to msdosfs_lock() in
> NetBSD and not in FreeBSD.  I don't know about Lites.  Linux has better
> msdosfs support in half as much code.

This is actually broken.  The underlying FS locking, as I said in a
previous message, should be moved to the vncalls layer (vfs_syscalls.c
in kern in BSD parlance).  The problem is in the use of vnode/inode
disassociation, which is broken anyway: the buffer chain for a vnode
will not be reclaimed even though it is in core if only a partial
chain exists.  The clustering code (unhappily) does not order the
buffers in a cluster in the LRU to be adjacent and in reverse order.

The Linux code uses non-vmio, so integration would require a lot of
work or using non-vmio.  The non-vmio in FreeBSD has not really been
kept up to date with the vmio code, so some integration from NetBSD
should be done if you choose this path.


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199511282130.OAA22102>