From owner-freebsd-current  Tue Feb 10 09:21:34 1998
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id JAA03314
          for current-outgoing; Tue, 10 Feb 1998 09:21:34 -0800 (PST)
          (envelope-from owner-freebsd-current@FreeBSD.ORG)
Received: from ns1.yes.no (ns1.yes.no [195.119.24.10])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA03260
          for <current@FreeBSD.ORG>; Tue, 10 Feb 1998 09:21:31 -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 RAA07786;
	Tue, 10 Feb 1998 17:21:25 GMT
Received: (from eivind@localhost)
	by bitbox.follo.net (8.8.6/8.8.6) id SAA18433;
	Tue, 10 Feb 1998 18:21:24 +0100 (MET)
Message-ID: <19980210182124.35246@follo.net>
Date: Tue, 10 Feb 1998 18:21:24 +0100
From: Eivind Eklund <eivind@yes.no>
To: Steve Passe <smp@csn.net>
Cc: Eivind Eklund <eivind@yes.no>, current@FreeBSD.ORG
Subject: Re: Minor test from somebody with SMP needed
References: <19980210174001.25447@follo.net> <199802101711.KAA23233@Ilsa.StevesCafe.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.88e
In-Reply-To: <199802101711.KAA23233@Ilsa.StevesCafe.com>; from Steve Passe on Tue, Feb 10, 1998 at 10:11:51AM -0700
Sender: owner-freebsd-current@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

On Tue, Feb 10, 1998 at 10:11:51AM -0700, Steve Passe wrote:
> Hi,
> 
> > Can somebody test if their system seem to work reliably with the below
> > patch and
> > 
> > options SIMPLELOCK_DEBUG 
> > 
> > in their kernel config file?  Alternatively, can somebody that
> > understand the issues say if the change is correct?
> > 
> > SIMPLELOCK_DEBUG doesn't work at all with a multiprocessor-system as
> > it is; the kernel won't even link, and this breaks LINT.
> 
> These routines can't work with SMP as they substitute for real locks
> and an SMP machine would soon go down in flames if it depended on them.
> They are a debug mechanism for a single CPU system, and predate the
> FreeBSD developmnent effort.

OK, what is the right fix, then?  I can see three obvious 'fixes' for
the time being:

1. Make SIMPLELOCK_DEBUG a no-op if NCPUS > 1

2. Rip out the SIMPLELOCK_DEBUG code (if it is no longer usable, and
not likely to become usable)

3. Do the below change (basically an error if compiling anything other
than LINT with SIMPLELOCK_DEBUG and SMP)

Index: kern_lock.c
===================================================================
RCS file: /home/ncvs/src/sys/kern/kern_lock.c,v
retrieving revision 1.16
diff -u -r1.16 kern_lock.c
--- kern_lock.c 1998/02/06 12:13:23     1.16
+++ kern_lock.c 1998/02/10 17:20:08
@@ -462,7 +462,7 @@
                printf(" with %d pending", lkp->lk_waitcount);
 }
 
-#if defined(SIMPLELOCK_DEBUG) && NCPUS == 1
+#if defined(SIMPLELOCK_DEBUG) && (NCPUS == 1 || defined(COMPILING_LINT))
 #include <sys/kernel.h>
 #include <sys/sysctl.h>
 
@@ -555,4 +555,6 @@
        if (curproc)
                curproc->p_simple_locks--;
 }
+#elif defined(SIMPLELOCK_DEBUG)
+#error "SIMPLELOCK_DEBUG is not compatible with SMP!"
 #endif /* SIMPLELOCK_DEBUG && NCPUS == 1 */


Just help me determine what is right, and I'll do the work.

Eivind.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe current" in the body of the message