Re: DEVFSv50 and /dev/fb? (or /dev/fb/? ???)

Richard Gooch (Richard.Gooch@atnf.CSIRO.AU)
Wed, 5 Aug 1998 18:17:27 +1000


Alan Cox writes:
> > I've heard the suggestion that you use initrd to populate /dev
> > automagically. IMHO that is simply silly. Firstly it takes time to
> > create all those inodes (for devices you have). When you shut down you
> > should probably remove those inodes. *This* is cleaner than devfs???
>
> Well its swappable, you can alter nodes on the fly, and it works. You
> are pretty much describing Solaris

What Solaris has is *really* ugly! About it being swappable: the RAM
consumption by devfs is quite small. A small system with few discs has
very few devfs inodes. A big system with lots of discs has
correspondingly more devfs inodes, but on the other hand such a system
is going to have lots of RAM. Besides, we're still only talking about
one or a few pages, even for a largish system.

> > If people want to automagically populate /dev from userspace, it
> > requires information from the kernel (current boot logs do not provide
> > sufficient information). Making the boot logs spit out information for
>
> It seems to all be there, and there are scsi tools to handle this. As
> a side item btw - a much more important trick to support would be
> "mount by uuid" you you can mount a volume by ID wherever it walks

And all the other devices we have in /dev? It's not just SCSI disc
inodes that pollute /dev.
BTW: I'm all for a UUID scheme too. Having UUID mounting doesn't
negate the need for devfs, though.

> [Note Im neither for or against devfs - its simply yet another random
> unapplied kernel patch]

But a good one, and one that should be applied! :-)

Regards,

Richard....

-
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