> Personally, I hope that ext3 won't bother with providing backwards
> compatibility to old kernels, but rather will clear out old backwards
> compatibility code, add the exciting new stuff (ACLs, capabilities)
> and fix some things (bigger [ug]id_t, dev_t, (off_t?)). The rest is
> missing the point, IMHO.
OK. If we make ext3 completely incompatible, so that it's not possible to use
2.0 kernels with it, then that removes the kernel compatibility question.
We can also note that tar _already_ doesn't cope with ext2 extended attributes,
and having it fail to handle caps is hardly the end of the world.
If we really want tar to work, then perhaps we should add (to libc) a function
which gets _all_ such things (extended attributes, ACLs, caps, access logs,...)
into a single block, which programs like tar and cp can handle transparently
(as if it were a resource fork), so they don't have to know about every
non-standard extension that we add to the inode data.
This is _not_ the same as adding a real resource fork in the filesystem, which
is generally acknowledged to be a bad thing. Think of it like an FTP daemon
which supports 'GET <directory>.tar.gz' to fetch a whole directory without
individually fetching every file contained therein.
My only remaining problem is NFS, but as NFS is widely acknowledged to be a
pile of poo anyway, my problem is more the fact that we haven't got a suitable
replacement for it yet, rather than that caps won't work over it.
In fact, I personally think caps not working over NFS would be a good thing,
because it may help to expedite its demise. (Coda-root, anyone?).
> Matthew - who went to the boat race on Saturday :(
hehe - I know I've already done this, but I can't resist the temptation to
gloat again in public :)
---- ---- ----
David Woodhouse David.Woodhouse@mvhi.com Office: (+44) 1223 810302
Project Leader, Process Information Systems Mobile: (+44) 976 658355
Axiom (Cambridge) Ltd., Swaffham Bulbeck, Cambridge, CB5 0NA, UK.
finger dwmw2@ferret.lmh.ox.ac.uk for PGP key.
-
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/