From owner-freebsd-smp  Thu Sep  5 16:46:19 1996
Return-Path: owner-smp
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id QAA28745
          for smp-outgoing; Thu, 5 Sep 1996 16:46:19 -0700 (PDT)
Received: from spinner.DIALix.COM (spinner.DIALix.COM [192.203.228.67])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA28724
          for <smp@freebsd.org>; Thu, 5 Sep 1996 16:46:11 -0700 (PDT)
Received: from spinner.DIALix.COM (localhost.DIALix.oz.au [127.0.0.1])
          by spinner.DIALix.COM (8.7.5/8.7.3) with ESMTP id HAA16950;
          Fri, 6 Sep 1996 07:43:29 +0800 (WST)
Message-Id: <199609052343.HAA16950@spinner.DIALix.COM>
X-Mailer: exmh version 1.6.7 5/3/96
To: Steve Passe <smp@csn.net>
cc: Terry Lambert <terry@lambert.org>, rv@groa.uct.ac.za, smp@freebsd.org
Subject: Re: SMP on Intel XXpress 
In-reply-to: Your message of "Thu, 05 Sep 1996 17:22:44 CST."
             <199609052322.RAA28908@clem.systemsix.com> 
Date: Fri, 06 Sep 1996 07:43:29 +0800
From: Peter Wemm <peter@spinner.dialix.com>
Sender: owner-smp@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Steve Passe wrote:
> Hi,
> 
> > after the "sysctl kern.smp_active=2" the 2nd CPU calls secondary_main()
> > which merely prints an info message, then calls cpu_switch(), from which
> > it never returns (thats a good thing, it shouldn't).  but from this point
> > on we have the INT loss problem.
> 
> is anything actual using the routine in i386/isa/vector.s:
> 
> FAST_INTR(irq_num, vec_name, enable_icus)
> 
> I believe it has a bug where it fails to release the mp_lock for
> hardware INTerrupts.

It would not suprise me if it was..  Something certainly is amis somewhere 
(although different to the problem that you're looking for), I'm almost 
sure I'm seeing effects of occasionally having both cpu's in the kernel at 
simultaniously somehow....  The current kernel mutex entry/exit is a 
nightmare to try and do locking for...

Cheers,
-Peter