"Theodore Y. Ts'o" wrote:
>
> Date: Thu, 20 Apr 2000 11:15:05 +0200 (MEST)
> From: R.E.Wolff@BitWizard.nl (Rogier Wolff)
>
[...]
>
> At first, the kernel publishing the attached devices sounds like a
> good idea. And a "filesystem" sounds like the obviously correct way to
> export that info. However, the permissions issue complicates it to the
> point where the "publish as a filesystem" policy should be
> reconsidered.
>
> That's basically right (as far as I'm concerned), although the daemon
> does have to be continuously running, to deal with hot-swap devices.
> It's the design I've always thought was the right one.
>
> People kept on saying that using a filesystem was the right approach,
> because the using a user-mode daemon "was too complicated", and then
> when people started pointing out shortcomings, the answer was "use
> devfsd". But after kludge after kludge started getting poured into
> devfsd (or into the VFS, which was Richard Gooch's latest attempt to
> work around this problem) to deal with the fact that there really is
> persistent state in /dev that needs to survive device insertion and
> removal, not to mention reboots ---- you have to start wondering which
> approach is really the more complicated one.
>
In my opinion this persistency problem is similar to what we have
with autoloading of kernel modules. Here we have information stored
in a config file (/etc/modules.conf), where you can set options which
get applied when a module is loaded by modprobe/insmod.
And devfsd works exactly the same! You can configure this information
in "/etc/devfsd.conf" (see man 5 devfsd)
Maybe the funcionality and/or syntax of devfsd and this configuration
file can be improved, but the basics are there already, and we use it
on our systems (Diskless Clients as well as "normal" Linux servers)
for months now without problems.
And I'm very much pro the extra (virtual) filesystem! Consider the
situation where you have /dev on a read-only root-fs! There you can't
just create new device special files on the fly and using a ramdisk
or writable NFS share just for the device files is a hack IMHO.
- andreas
-- Andreas Haumer | email: andreas@xss.co.at | http://www.xss.co.at *x Software + Systeme | phone: +43.1.6060114-0 | Karmarschgasse 51/2/20 | +43.664.3004449 | A-1100 Vienna, Austria | fax: +43.1.6060114-71 | AH327-RIPE- 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 : Sun Apr 23 2000 - 21:00:17 EST