On Mon, 21 Feb 2000, Tony Hoyle wrote:
>I had the idea of the 'virtual mount'. Basically you mount with -o virtual, and the vfs records
>the mount details, then returns immediately. When the first non-stat operation is done on the mount
>point, the mount actually happens.
Let me be the voice of reason here:
Why the hell would you want to do that? <grin>
How can you be sure the file system is mountable until you actually mount it?
What's the advantage of doing something like this?
>I mostly implemented it last night... however I thought I'd better put my ideas down here so I can be shot down in
>flames (or even encouraged) for having the idea in the first place. (It turned out to be much simpler than I thought,
>although I haven't acctually attempted the mount-from-vfs yet...)
It's good kernel hacking exercise... but I'd have to rank this up there
with devfs in stupidity (no offense.)
--Ricky
PS: I stand by my original (over a year old) stance that devfs is a
"Bad Idea (tm)". It could be turned into a good idea... I'm more
inclined to use something like solaris uses.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Feb 23 2000 - 21:00:28 EST