From owner-freebsd-hackers  Sun Dec 14 11:23:00 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id LAA01891
          for hackers-outgoing; Sun, 14 Dec 1997 11:23:00 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from ns1.yes.no (ns1.yes.no [195.119.24.10])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA01880
          for <freebsd-hackers@FreeBSD.ORG>; Sun, 14 Dec 1997 11:22:53 -0800 (PST)
          (envelope-from eivind@bitbox.follo.net)
Received: from bitbox.follo.net (bitbox.follo.net [194.198.43.36])
	by ns1.yes.no (8.8.7/8.8.7) with ESMTP id TAA10791
	for <freebsd-hackers@FreeBSD.ORG>; Sun, 14 Dec 1997 19:22:51 GMT
Received: (from eivind@localhost)
	by bitbox.follo.net (8.8.6/8.8.6) id UAA13250;
	Sun, 14 Dec 1997 20:22:50 +0100 (MET)
X-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch)
To: freebsd-hackers@FreeBSD.ORG
Subject: Re: New option model (was Re: cvs commit: src/sys/kern kern_exit.c)
References: <199712071816.KAA21840@freefall.freebsd.org>     <19971208091729.40125@uriah.heep.sax.de>     <863ek4p1tx.fsf_-_@bitbox.follo.net>     <199712100013.BAA04827@uriah.heep.sax.de>     <863ek1xtsl.fsf@bitbox.follo.net> <199712141816.TAA29536@uriah.heep.sax.de>
From: Eivind Eklund <perhaps@yes.no>
Date: 14 Dec 1997 20:22:48 +0100
In-Reply-To: j@uriah.heep.sax.de's message of Sun, 14 Dec 1997 19:16:08 +0100 (MET)
Message-ID: <86zpm3pw53.fsf@bitbox.follo.net>
Lines: 38
X-Mailer: Gnus v5.4.52/XEmacs 20.2
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

j@uriah.heep.sax.de (J Wunsch) writes:

> Eivind Eklund <perhaps@yes.no> wrote:
> 
> > As far as I can see, the only correct #ifdef's in the kernel .c files
> > are #ifdef KERNEL (should be _KERNEL) - everything else should be
> > #if to allow #define XXX 1 or 0.  Am I correct?  Should I fix this as
> > I do my walkthrough anyway?
> 
> Hmm, not sure.  All the boolean current options are #ifdef-evaluated.
> 
> What do others think about it?  I don't see an urgent need to change
> _this_.

I didn't notice all of those options had an 'N' prefixed :-(
NPPP_DEFLATE still seems somehow bogus.

Fixing the non-header file versions would then just be adding the
proper opt_xxx.h includes - should be trivial, except I'm not certain
what kernel code is being re-vamped at any moment.

> Well, if you were going to spell _KERNEL correctly, by all means, go
> ahead! ;)

Won't this break many patches people are keeping externally?

> I'm not sure what others think about kernel structs that vary their
> size based on different kernel option usage.  IMHO they should die,
> since they are highly dangerous, and this danger doesn't justify the
> few saved bytes.

I dislike them, at least.

<DIVERSION>I also dislike using structs to communicate kernel <->
userland - e.g. for ipfw.  I'd like this to use something like the
tags system on the Amiga.</DIVERSION>

Eivind.