From owner-cvs-sys  Wed Nov  8 00:30:39 1995
Return-Path: owner-cvs-sys
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id AAA27199
          for cvs-sys-outgoing; Wed, 8 Nov 1995 00:30:39 -0800
Received: from tfs.com (tfs.com [140.145.250.1])
          by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id AAA27193
          ; Wed, 8 Nov 1995 00:30:31 -0800
Received: from critter.tfs.com by tfs.com (smail3.1.28.1) with SMTP
	id m0tD5tu-0003vwC; Wed, 8 Nov 95 00:30 PST
Received: from localhost (localhost [127.0.0.1]) by critter.tfs.com (8.6.11/8.6.9) with SMTP id JAA01074; Wed, 8 Nov 1995 09:30:28 +0100
X-Authentication-Warning: critter.tfs.com: Host localhost didn't use HELO protocol
To: CVS-commiters@freefall.freebsd.org, cvs-sys@freefall.freebsd.org
Subject: Re: cvs commit: src/sys/kern kern_sysctl.c 
In-reply-to: Your message of "Wed, 08 Nov 1995 04:41:50 +1100."
             <199511071741.EAA12200@godzilla.zeta.org.au> 
Date: Wed, 08 Nov 1995 09:30:27 +0100
Message-ID: <1072.815819427@critter.tfs.com>
From: Poul-Henning Kamp <phk@critter.tfs.com>
Sender: owner-cvs-sys@FreeBSD.org
Precedence: bulk


Bruce I share most of your concerns, but disagree on the cure.

Some specifics:

Yes, I rely on compiler magic for the sysctl variables.  I live in the
nineties, and we have to move forward.  I looked at other options, and
they were uglier.

#ifdef TERRY_MODE
The only thing we could possibly do which would be a benefit, would be
to make an intelligent FreeBSD preprocessor that ran over all the
sources first and did some of all the grunt work we presently overload 
cpp with.

This would for instance allow us to split data into two classes, one 
of which could be returned to malloc/free when it wasn't used anymore,
allow us to make pinned/unpinned sections of drivers so that we could
page part of the kernel (!) and so on...

#endif /* TERRY_MODE */

I have included a string for description of the variables, which is
presently not instatiated in the kernel.  My idea was to have a program
do a 
	find /usr/src/sys -type f -print | xargs gronk_out_sysctl_info > foo
and proceed from there.

It is overkill to expand the (possibly rather lengthy) descriptions into
the kernel unless we want to have them available in the user-config, and
the descriptions belong in the source, with the definition of the variable.

I do plan to (have somebody else) add  sysctl-variables to the userconfig
editor.

I anticipate a "bulk" operator for sysctl, either working along the lines
of readv(2), SNMP's getbulk or by allowing one to fetch entire subtrees at 
a time, like "kern.vm.*".

Yes, I would really love to kill the "struct" variables in the sense of
everything interesting in them be available as simple variables too.
But I still think we need three types: Int, Text, and Opaque.
I don't like the concept of converting big structures to ascii and 
back again, and there are some legitimate uses of entire structures.
Maybe we should simply adopt the SNMP types without the ASN.1 encoding.

I would also love to convert from the int[] names to "char *" at some
later time.

For now, I'm only working on step one, replace the current definition-
method with something more flexible, comprehensive and usable.

Functional enhancements will follow later.

--
Poul-Henning Kamp           | phk@FreeBSD.ORG       FreeBSD Core-team.
http://www.freebsd.org/~phk | phk@login.dknet.dk    Private mailbox.
whois: [PHK]                | phk@ref.tfs.com       TRW Financial Systems, Inc.
Future will arrive by its own means, progress not so.