Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 14 Jul 2000 11:21:17 +0200
From:      Adrian Chadd <adrian@FreeBSD.ORG>
To:        Matthew Dillon <dillon@apollo.backplane.com>
Cc:        Adrian Chadd <adrian@FreeBSD.ORG>, Andrzej Bialecki <abial@webgiro.com>, freebsd-arch@FreeBSD.ORG
Subject:   Re: SysctlFS
Message-ID:  <20000714112117.D17372@ywing.creative.net.au>
In-Reply-To: <200007130537.WAA29614@apollo.backplane.com>; from dillon@apollo.backplane.com on Wed, Jul 12, 2000 at 10:37:50PM -0700
References:  <Pine.BSF.4.20.0007121328020.49102-100000@mx.webgiro.com> <20000712144510.A11316@ywing.creative.net.au> <200007130537.WAA29614@apollo.backplane.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jul 12, 2000, Matthew Dillon wrote:
> 
> :...
> :> to the names should still be retrieved in binary form, as they are
> :> exported via SYSCTL_* macros. But filesystem paradigm would allow us to
> :> reuse all the concepts for hierarchical name handling, traversal,
> :> permissions etc... The sysctlFS nodes would be probably read-only from
> :> userland, as I don't see much sense in userland programs renaming or
> :> removing them - they would be created, named and removed from
> :> kernel-land. But things like traversal and access would be simplified
> :> greatly.
> :> 
> :> Any thoughts?
> :
> :I'm probably going to poke at it in a few weeks as an "example filesystem"
> :for some documentation I'm writing up. There are issues in having it as
> :a filesystem - see how /proc needs to be handled for jails right now. 
> :I'm sure other people on the list can fill you in .. :)
> :
> :
> :Adrian
> :
> :-- 
> :Adrian Chadd			Build a man a fire, and he's warm for the
> :<adrian@FreeBSD.org>		rest of the evening. Set a man on fire and
> 
>     I will point out that Linux puts system config variables in /proc and
>     it has been an unmitigated disaster.  Well, maybe not *that bad*, but
>     it's fairly obvious to me that putting the config variables in a 
>     filesystem yields absolutely *NO* advantage over having a system call
>     (and /sbin util) to do it.
> 
>     The current sysctl methodology works just dandy, we should not mess with
>     it.

Oh, I agree. Hence why I said I'd write it up as an "example filesystem" ..
I'm tossing up a few candidates at the moment.

-- 
Adrian Chadd			Build a man a fire, and he's warm for the
<adrian@FreeBSD.org>		rest of the evening. Set a man on fire and
				he's warm for the rest of his life.



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000714112117.D17372>