From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755224Ab1CYLWK (ORCPT ); Fri, 25 Mar 2011 07:22:10 -0400 Received: from ozlabs.org ([203.10.76.45]:58703 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753564Ab1CYLWE (ORCPT ); Fri, 25 Mar 2011 07:22:04 -0400 From: Rusty Russell To: Christian Borntraeger , Anthony Liguori Cc: Christoph Hellwig , qemu-devel@nongnu.org, linux-kernel@vger.kernel.org, stefanha@gmail.com, kwolf@redhat.com, prerna@linux.vnet.ibm.com Subject: Re: [PATCH, RFC] virtio_blk: add cache control support In-Reply-To: <4D8B14BD.2060602@de.ibm.com> References: <20110315141049.GA30627@lst.de> <20110315141644.GA30803@lst.de> <87y64fhfjw.fsf@rustcorp.com.au> <20110316140958.GB21877@lst.de> <877hbygwu7.fsf@rustcorp.com.au> <4D8AB514.5020306@us.ibm.com> <4D8B14BD.2060602@de.ibm.com> User-Agent: Notmuch/0.3.1 (http://notmuchmail.org) Emacs/23.1.1 (i686-pc-linux-gnu) Date: Fri, 25 Mar 2011 15:38:51 +1030 Message-ID: <8762r7lrbw.fsf@rustcorp.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 24 Mar 2011 10:54:05 +0100, Christian Borntraeger wrote: > Am 24.03.2011 04:05, schrieb Anthony Liguori: > >> ie. lguest and S/390 don't trap writes to config space. > >> > >> Or perhaps they should? But we should be explicit about needing it... > > I don't think we ever operated on the assumption that config space writes would trap. > > > > I don't think adding it is the right thing either because you can do byte access to the config space which makes atomicity difficult. > > There is the additional problem, that s390 has no MMIO and,therefore, > there is no real HW support for trapping writes to an area. You can > use page faults, or read-only faults on newer systems, but this is > expensive. In addition, page faults only deliver the page frame, but > not the offset within a page. That's not *really* a problem, since you have control over the config_set operation and could do whatever you wanted. But I wanted to make sure we're all on the same page: you *can't* rely on the host knowing immediately what you write to the config space. If you want that, an actual queued request is necessary... Thanks, Rusty.