From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner+w=401wt.eu-S934427AbXHLLVj@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S934427AbXHLLVj (ORCPT <rfc822;w@1wt.eu>);
	Sun, 12 Aug 2007 07:21:39 -0400
Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755917AbXHLLVc
	(ORCPT <rfc822;linux-kernel-outgoing>);
	Sun, 12 Aug 2007 07:21:32 -0400
Received: from adsl-67-117-79-109.dsl.sntc01.pacbell.net ([67.117.79.109]:3379
	"EHLO aurum.uhlenkott.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1760128AbXHLLVb (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Sun, 12 Aug 2007 07:21:31 -0400
X-Greylist: delayed 332 seconds by postgrey-1.27 at vger.kernel.org; Sun, 12 Aug 2007 07:21:31 EDT
Date: Sun, 12 Aug 2007 04:15:58 -0700
From: Jason Uhlenkott <jasonuhl@jasonuhl.org>
To: Willy Tarreau <w@1wt.eu>
Cc: linux-kernel@vger.kernel.org, stable@kernel.org,
       Thomas Gleixner <tglx@linutronix.de>, john stultz <johnstul@us.ibm.com>,
       Dave Jones <davej@redhat.com>, Ingo Molnar <mingo@elte.hu>,
       Vincent Fortier <Vincent.Fortier1@EC.GC.CA>,
       Chris Wright <chrisw@sous-sol.org>, Greg Kroah-Hartman <gregkh@suse.de>
Subject: Re: [2.6.20.16 review 27/28] NTP: remove clock_was_set() call to prevent deadlock
Message-ID: <20070812111558.GA18647@aurum.uhlenkott.net>
References: <20070811184752.%N@1wt.eu> <20070811184857.%N@1wt.eu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
In-Reply-To: <20070811184857.%N@1wt.eu>
User-Agent: Mutt/1.5.9i
Sender: linux-kernel-owner@vger.kernel.org
X-Mailing-List: linux-kernel@vger.kernel.org

On Sat, Aug 11, 2007 at 21:48:19 +0200, Willy Tarreau wrote:
> The clock_was_set() call in seconds_overflow() which happens only when
> leap seconds are inserted / deleted is wrong in two aspects:
> 
> 1. it results in a call to on_each_cpu() with interrupts disabled
> 2. it is potential deadlock source vs. call_lock in smp_call_function()

clock_was_set() is a no-op in 2.6.20, so this one looks unnecessary
(but harmless).  Thankfully the "hang every Linux box on the planet
simultaneously" regression (okay, that's *slight* hyperbole) was
limited to 2.6.21.