Re: [Xen-devel] page ref/type count overflows

To:	Jan Beulich <jbeulich@xxxxxxxxxx>, <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject:	Re: [Xen-devel] page ref/type count overflows
From:	Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
Date:	Mon, 26 Jan 2009 13:51:12 +0000
Cc:
Delivery-date:	Mon, 26 Jan 2009 05:51:34 -0800
Envelope-to:	www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to:	<C5A36C45.1BDC%keir.fraser@xxxxxxxxxxxxx>
List-help:	<mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id:	Xen developer discussion <xen-devel.lists.xensource.com>
List-post:	<mailto:xen-devel@lists.xensource.com>
List-subscribe:	<http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe:	<http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender:	xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index:	Acl/ur0ZUqRKrWkdnU2n7WXpz2EAfAAAmjro
Thread-topic:	[Xen-devel] page ref/type count overflows
User-agent:	Microsoft-Entourage/12.14.0.081024

On 26/01/2009 13:33, "Keir Fraser" <keir.fraser@xxxxxxxxxxxxx> wrote:

> As for the general issue, why fudge around it when for any modern system we
> can just fix it? By which I mean, x64 systems with CMPXCHG16 support
> (everything but ancient Opterons?) can have a full domain pointer and a long
> count_info in struct page_info, and still update both atomically.

Actually I think we could do without CX16. I'm not sure count_info and
domain need to be checked/updated at the same time. I think it would suffice
to update count_info, then check domain and undo on mismatch. That'd get rid
of the existing ugly open-coded cmpxchg8b operations too.

I think I'll look into this bit at least...

 -- Keir

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel

WARNING - OLD ARCHIVES

xen-devel

Re: [Xen-devel] page ref/type count overflows