> It's amusing to note that our ext2fs implementation gives you the 
> traditional FFS guarantee about metadata, *unlike the Linux implementation*,
> so you can actually use it with more confidence than you can use the native
> ext2fs in Linux.  The downside is that it's a bit slower, but that's because
> it actually does the right thing if the system crashes, instead of potentially
> eating your filesystem itself.

Just to make sure I have a clear understanding.. is the following table
accurate?

os	mount option		data	meta-data
------	----------------------	------	---------------
netbsd	normal (default)	async	sync
	sync			sync	sync
	async			async	async
	softdep			async	"softdep"
linux	async (default)		async	async
	sync			sync	sync


From slackware 8.0 mount(8):
              async  All  I/O  to  the file system should be done
		     asynchronously.
              sync   All I/O to the file system  should  be  done
		     synchronously.

From netbsd 1.5.2 mount(8):
             async       All I/O to the file system should be done asyn-
		         chronously.
             sync        All I/O to the file system should be done syn-
		         chronously. This is not equivalent to the normal mode
			 in which only metadata is written synchronously.
             softdep     (FFS only) Mount the filesystem using soft-dependen-
                         cies. This means that metadata will not be written
                         immediately, but is written in an ordered fashion to
                         keep the on-disk state of the filesystem consistent.
Ben
-- 
NetBSD: Abandon all hype, oh ye who enter here.