From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [PATCH 15/15] KVM: Fix racy in kvm_free_assigned_irq
Date: Sat, 27 Dec 2008 18:06:26 -0200
Message-ID: <20081227200626.GA4095@amt.cnet>
References: <20081225115609.GA10087@syang10-desktop> <1230258607-15208-1-git-send-email-sheng@linux.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Avi Kivity <avi@redhat.com>, kvm@vger.kernel.org
To: Sheng Yang <sheng@linux.intel.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx2.redhat.com ([66.187.237.31]:41127 "EHLO mx2.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1753286AbYL0UGn (ORCPT <rfc822;kvm@vger.kernel.org>);
	Sat, 27 Dec 2008 15:06:43 -0500
Content-Disposition: inline
In-Reply-To: <1230258607-15208-1-git-send-email-sheng@linux.intel.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Fri, Dec 26, 2008 at 10:30:07AM +0800, Sheng Yang wrote:
> Thanks to Marcelo's observation, The following code have potential issue:
> 
> if (cancel_work_sync(&assigned_dev->interrupt_work))
> 	kvm_put_kvm(kvm);
> 
> In fact, cancel_work_sync() would return true either work struct is only
> scheduled or the callback of work struct is executed. This code only
> consider the former situation.

Why not simply drop the reference inc / dec from irq handler/work
function?

Just make sure that there is no queued/executing work left behind on
vm shutdown. Don't think an additional reference is necessary. Or am I
missing something?

> Also, we have a window between cancel_work_sync() and free_irq. 

This one looks OK.