Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753598AbYAWPsx (ORCPT ); Wed, 23 Jan 2008 10:48:53 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753652AbYAWPs3 (ORCPT ); Wed, 23 Jan 2008 10:48:29 -0500 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:50913 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753607AbYAWPs1 (ORCPT ); Wed, 23 Jan 2008 10:48:27 -0500 Date: Wed, 23 Jan 2008 09:48:23 -0600 From: Robin Holt To: Gerd Hoffmann Cc: Robin Holt , Christoph Lameter , Andrea Arcangeli , Andrew Morton , Nick Piggin , linux-mm@kvack.org, Benjamin Herrenschmidt , steiner@sgi.com, linux-kernel@vger.kernel.org, Avi Kivity , kvm-devel@lists.sourceforge.net, daniel.blueman@quadrics.com, Hugh Dickins Subject: Re: [kvm-devel] [PATCH] export notifier #1 Message-ID: <20080123154823.GG3058@sgi.com> References: <20080121125204.GJ6970@v2.random> <4795F9D2.1050503@qumranet.com> <20080122144332.GE7331@v2.random> <20080122200858.GB15848@v2.random> <4797384B.7080200@redhat.com> <20080123131939.GJ26420@sgi.com> <47974B54.30407@redhat.com> <20080123141814.GE3058@sgi.com> <479750CA.4070101@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <479750CA.4070101@redhat.com> User-Agent: Mutt/1.5.15+20070412 (2007-04-11) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1276 Lines: 28 On Wed, Jan 23, 2008 at 03:35:54PM +0100, Gerd Hoffmann wrote: > Robin Holt wrote: > > We have a seg structure which is similar to some structure you probably > > have which describes the grant. One of the things hanging off that > > seg structure is essentially a page table containing PFNs with their > > respective flags (XPMEM specific and not the same as the pfn flags in > > the processor page tables). > > i.e. page tables used by hardware != cpu, right? Actually page tables used exclusively by software during the cross partition coordination. Those entries are inserted on the remote side by normal faults with VM_PFNMAP vmas created by the importing side. > In the Xen guest case the normal processor page tables are modified, but > in a special way to make the Xen hypervisor also release the grant. In our guest case, we can not access the kernel struct page area on the remote host. We therefore handle all the ref/deref of the page as part of messaging the PFN across the partition boundaries. Thanks, Robin -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/