Mouse writes:

>_I_ think that if NetBSD's NFS implementation doesn't work, it should
>be removed.  Whether it's good or not, if it doesn't work, it shouldn't
>pretend to.

I think this misses the point completely.

The NFS protocol implementation is fairly solid.  After all, it's
based on Rick Macklem's code, which is in 4.4bsd and FreeBSD as well.
As I understand it, the problem is lack of consistency between
the buffer and VM caches.  (Or something.)

>But the problem is worse than that, as I understand it; it's not just
>mmap(), but anything that pages to a file.

Precisely, which is why talking about using AFS instead misses the point, too.
(Besides, personally, I find the costs of using AFS not worth the
hassle in an environment smaller than 1000's of users on 1000's of hosts.)