From mboxrd@z Thu Jan  1 00:00:00 1970
From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Subject: Re: [PATCH 5/5] AMD IOMMU: widen NUMA nodes to be
 allocated from
Date: Tue, 10 Mar 2015 09:55:47 -0400
Message-ID: <54FEF7E3.4000302@oracle.com>
References: <54EF315902000078000640FF@mail.emea.novell.com>	<54EF341C020000780006416C@mail.emea.novell.com>	<54F892CD.2000300@citrix.com>	<54F96A440200007800066D57@mail.emea.novell.com>	<54F99A63.6040200@citrix.com>
	<54FDBF73.8040407@amd.com>	<54FDD7C4.5020905@citrix.com>
	<54FDEE37.7020008@amd.com>
	<54FEACEE0200007800067E1F@mail.emea.novell.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; Format="flowed"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
Received: from mail6.bemta3.messagelabs.com ([195.245.230.39])
	by lists.xen.org with esmtp (Exim 4.72)
	(envelope-from <boris.ostrovsky@oracle.com>) id 1YVKex-0006C4-UA
	for xen-devel@lists.xenproject.org; Tue, 10 Mar 2015 13:57:24 +0000
In-Reply-To: <54FEACEE0200007800067E1F@mail.emea.novell.com>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: Jan Beulich <JBeulich@suse.com>, Suravee Suthikulanit <suravee.suthikulpanit@amd.com>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>, Dario Faggioli <dario.faggioli@citrix.com>, Aravind Gopalakrishnan <aravind.gopalakrishnan@amd.com>, xen-devel <xen-devel@lists.xenproject.org>
List-Id: xen-devel@lists.xenproject.org

On 03/10/2015 03:35 AM, Jan Beulich wrote:
>>>> On 09.03.15 at 20:02, <suravee.suthikulpanit@amd.com> wrote:
>> I agree that having the IO page tables on the NUMA node that is closest
>> to the IOMMU would be beneficial.
> And I already withdrew this patch and the corresponding VT-d one.
>
>> However, I am not sure at the moment
>> that this information could be easily determined. I think ACPI _PXM for
>> devices should be able to provide this information, but this is optional
>> and often not available.
> And even if it was available, it would be too late at least for Dom0's
> allocations (as it requires Dom0's interpreter to dig out this detail).
> The best we could do in that case would be to try to replace the
> existing tables. Or assume Dom0 is being placed suitably by the
> dom0_nodes= option. Or add yet another option.

There is a nodeID register on each northbridge (D1<x>F0x60). You would 
have to figure out how to map it to _PXMs though.

-boris