SCT> No good: as soon as you export such a directory over NFS to a
SCT> non-Linux box, the client will not be able to read the data.
SCT> Systems like Gnome and KDE are not restricted to running on Linux
SCT> only.
It would not work if the client did not support forks in the way I
proposed. It would however work to have the server not support forks
as long as the client does support it.
If you really wanted you could wrap the read() in user space on the
client that does not support forks, so that it attempts to read
foo/default before returning EISDIR on foo. That way my proposal
works without kernel help, albeit less efficiently.
Besides, NFS is not even Posix compliant. If we use NFS as a design
standard for future file systems, nothing new will ever happen.
Benny
-
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.altern.org/andrebalsa/doc/lkml-faq.html