From owner-freebsd-hackers  Tue Oct  1 09:32:28 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id JAA28829
          for hackers-outgoing; Tue, 1 Oct 1996 09:32:28 -0700 (PDT)
Received: from lestat.nas.nasa.gov (lestat.nas.nasa.gov [129.99.50.29])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id JAA28820;
          Tue, 1 Oct 1996 09:32:22 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by lestat.nas.nasa.gov (8.7.5/8.6.12) with SMTP id JAA01098; Tue, 1 Oct 1996 09:21:53 -0700 (PDT)
Message-Id: <199610011621.JAA01098@lestat.nas.nasa.gov>
X-Authentication-Warning: lestat.nas.nasa.gov: Host localhost [127.0.0.1] didn't use HELO protocol
To: Bruce Evans <bde@zeta.org.au>
Cc: dyson@freebsd.org, jehamby@lightside.com, chuckr@glue.umd.edu,
        hackers@freebsd.org, karl@mcs.net
Subject: Re: PS broke again -- what has to be rebuilt to stop this? 
Reply-To: Jason Thorpe <thorpej@nas.nasa.gov>
From: Jason Thorpe <thorpej@nas.nasa.gov>
Date: Tue, 01 Oct 1996 09:21:48 -0700
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Tue, 1 Oct 1996 18:54:46 +1000 
 Bruce Evans <bde@zeta.org.au> wrote:

 > I think the main advantage of kernfs is that it is more efficient in
 > shell scripts.  Running sysctl(8) to fetch a value from the kernel
 > and format it nicely takes a while.

Actually, if I remember right, the whole point of kernfs was really
just an example of something cool you could do with the 4.4BSD vfs code.

(Keeping in mind I'm talking about NetBSD here) I don't usually
rely on it for _anything_ (even in shell scripts, though a port or
two use it for access to the msgbuf .. saves having to include dmesg
in the miniroot :-) because I'm not willing to tell a user "you have
to have this in your kernel, or else you will lose."

[ going back to what the original topic seems to have been... ]

In my mind, using /proc for ps(1) is just silly... if you're reading
structures, you still have the same problem as you have now, if
you're reading and parsing strings (more "portable"), it gets slow.
Besides, you have to keep the kvm code in ps(1) anyhow, else you lose
the ability to analyze crashdumps with it.  The marginal gain doesn't
seem worth the clutter.

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939