From owner-freebsd-hackers  Fri Jul 12 14:16:11 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id OAA16850
          for hackers-outgoing; Fri, 12 Jul 1996 14:16:11 -0700 (PDT)
Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id OAA16841;
          Fri, 12 Jul 1996 14:16:06 -0700 (PDT)
Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id OAA01986; Fri, 12 Jul 1996 14:10:47 -0700
From: Terry Lambert <terry@lambert.org>
Message-Id: <199607122110.OAA01986@phaeton.artisoft.com>
Subject: Re: Preach it (was Some recent changes to GENERIC)
To: phk@FreeBSD.ORG (Poul-Henning Kamp)
Date: Fri, 12 Jul 1996 14:10:47 -0700 (MST)
Cc: pjchilds@imforei.apana.org.au, terry@lambert.org,
        freebsd-hackers@FreeBSD.ORG
In-Reply-To: <3731.837161036@critter.tfs.com> from "Poul-Henning Kamp" at Jul 12, 96 10:43:56 am
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> I agree, how about this for a revolution:
> 
> I have this idea about putting tcl in the kernel.  I can imagine some
> rather interesting possibilities this would give us.
> 
> Imagine all the "policies" we have, they could be boot-time configurable.
> 
> When we run out of vm for instance, If we had tcl in the
> kernel the sysad could do something like:
> 
> proc out_of_vm {} {
> 	foreach p in [procs] {
> 		if {[proc argv0 $p] == "emacs"} {
> 			proc kill -6 $p
> 		}
> 	}
> }
> 
> I'm almost kidding :-)

I will support this, if the code is paged in for use, paged out when
no longer in use, dynamically configurable into the machine without
taking it down, and seperately installed.  This means seperating
rc into multiple files in a directory in order to enable "drop in"
install, and/or kernel-initated module loading without the current
relink interface (which I prototyped, so I know it is crud).

I have no problem with "zero-overhead-unless-used" implementations of
*anything*.


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