Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 12 Jan 1998 19:32:10 +0000 (GMT)
From:      Terry Lambert <tlambert@primenet.com>
To:        brian@awfulhak.org (Brian Somers)
Cc:        freebsd-hackers@FreeBSD.ORG
Subject:   Re: Change to config(8)
Message-ID:  <199801121932.MAA27920@usr04.primenet.com>
In-Reply-To: <199801110342.DAA20820@awfulhak.demon.co.uk> from "Brian Somers" at Jan 11, 98 03:42:29 am

next in thread | previous in thread | raw e-mail | index | archive | help
> I'd like to commit this change, but I don't know if it's correct.
> 
> AFAICT, because the minor() bit of a dev_t is only 8 bits, minors of >
> 256 are useless.  They also break the -current DEVFS code in that you 
> end up with an increased major number (oops!) if you (for example)
> 
>   pseudo-device tun 257
> 
> However, I haven't committed it because I don't know where things 
> like /dev/sd0s1 get their minors (0x00020002).

Too much crap is encoded in minors.  That's the main reason for DEVFS,
in my mind.  You *can't* netboot a FreeBSD box from a non-FreeBSD
box in most cases because you *can't* mount a /dev with large enough
minor's in the mknod for a physical /dev in the NFS mounted root.

IMO, I think "tun" and other devices should be clone devices; this
would allow you to dynamically allocate them without having to
specify a hard count.

You could still enforce a quota on instances, assuming you wanted to
for whatever reason unknown to me...


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199801121932.MAA27920>