From owner-freebsd-hackers  Sat Aug  3 14:04:39 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id OAA14993
          for hackers-outgoing; Sat, 3 Aug 1996 14:04:39 -0700 (PDT)
Received: from clem.systemsix.com (clem.systemsix.com [198.99.86.131])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id OAA14844;
          Sat, 3 Aug 1996 14:02:50 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by clem.systemsix.com (8.6.12/8.6.12) with SMTP id PAA14656; Sat, 3 Aug 1996 15:02:40 -0600
Message-Id: <199608032102.PAA14656@clem.systemsix.com>
X-Authentication-Warning: clem.systemsix.com: Host localhost didn't use HELO protocol
X-Mailer: exmh version 1.6.5 12/11/95
From: Steve Passe <smp@csn.net>
To: smp@freebsd.org
cc: hackers@freebsd.org
Subject: halt & sysctl
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sat, 03 Aug 1996 15:02:40 -0600
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Hi,

I have observed that when halting SMP kernels the 'sync' never
finishes unless I manually disable the 2nd CPU via:

sysctl -w kern.smp_active=0

before issuing the halt command.  I have been looking for a way
to do the equivalent in machdep.c:boot():

boot(howto)
	int howto;
{
#ifdef SMP
/**
 * FIXME: all but the first CPU needs to be disabled here!!!
 *        we want to do equiv. of "sysctl -w kern.smp_active=0"
 */
       int status;
       int sctl[2];

	sctl[0] = CTL_KERN;
>>>    sctl[1] = ???;
       status = kernel_sysctl(&proc0, sctl, 2, 0, 0, 0, sizeof( int ), 0);
       printf( "\nsysctl status: %d\n", status );
#endif /* SMP */

	if (!cold && (howto & RB_NOSYNC) == 0 && waittime < 0) {
...
}

Am I barking up the right tree?
Anyone know the proper value for sctl[1]?

--
Steve Passe	| powered by
smp@csn.net	|            FreeBSD