From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48141) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YLXJg-00083X-9a for qemu-devel@nongnu.org; Wed, 11 Feb 2015 08:26:57 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1YLXJc-0007Ea-8l for qemu-devel@nongnu.org; Wed, 11 Feb 2015 08:26:56 -0500 Received: from mx1.redhat.com ([209.132.183.28]:56447) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YLXJc-0007Bi-1J for qemu-devel@nongnu.org; Wed, 11 Feb 2015 08:26:52 -0500 Message-ID: <54DB5894.7090501@redhat.com> Date: Wed, 11 Feb 2015 14:26:44 +0100 From: Paolo Bonzini MIME-Version: 1.0 References: <1423572769-4238-1-git-send-email-pbonzini@redhat.com> <20150211051302.3809.50882@loki> <54DB04F8.3080405@redhat.com> <20150211131641.3809.42730@loki> In-Reply-To: <20150211131641.3809.42730@loki> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [PATCH] memory: unregister AddressSpace MemoryListener within BQL List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Michael Roth , qemu-devel On 11/02/2015 14:16, Michael Roth wrote: > Since the AddressSpace is a fields of PCIDevice, do we maybe need to make > sure it's refcount doesn't drop to 0 prior to the RCU callback taking > place? Yes, but then the problem is that objects are finalized outside the big QEMU lock. This is a pretty large can of worms, because finalizing an object can in turn finalize other objects, which means a lot of mutexes have to be introduced. I'll think a bit more about it. Paolo