Re: [Xen-devel] Testing nested virtualization on Intel CPUs

To:	Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>, Tim Deegan <tim@xxxxxxx>
Subject:	Re: [Xen-devel] Testing nested virtualization on Intel CPUs
From:	Keir Fraser <keir.xen@xxxxxxxxx>
Date:	Wed, 21 Sep 2011 10:23:16 -0700
Cc:	xen-devel@xxxxxxxxxxxxxxxxxxx, Eddie Dong <eddie.dong@xxxxxxxxx>
Delivery-date:	Wed, 21 Sep 2011 10:23:52 -0700
Dkim-signature:	v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=user-agent:date:subject:from:to:cc:message-id:thread-topic :thread-index:in-reply-to:mime-version:content-type :content-transfer-encoding; bh=2v25S6mDxNM8g6ZczWX639bIOAzvPZ4n+B1KubCDjgE=; b=fsFDub2ER0zpPKMWVPymL7BTj1tXe1tKHhOCHfG8l8yM66D/5GFPYzAu2AyfYwfGNS 7Lwde+8SGwqucVglpheMpEmr5Olk3N5NrcG9ei3y8MC4c4NdMZ9UOMP0xlOCgPGToMFy 0ZLqpfAgwS+xMbQ5IlPcOh7pQdoerz8dZSuWk=
Envelope-to:	www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to:	<20110921175039.4ddcbb39@xxxxxxxxxxxxxxxxxxx>
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:	Acx4gyX4ZpgK8pgio0y0PVqvYK6BSA==
Thread-topic:	[Xen-devel] Testing nested virtualization on Intel CPUs
User-agent:	Microsoft-Entourage/12.30.0.110427

On 21/09/2011 09:50, "Alan Cox" <alan@xxxxxxxxxxxxxxxxxxx> wrote:

> On Wed, 21 Sep 2011 17:38:10 +0100
> Tim Deegan <tim@xxxxxxx> wrote:
> 
>> At 22:18 -0700 on 20 Sep (1316557136), AP wrote:
>>>> Yes, booting Xen as a nested guest is very slow at startup, because of
>>>> how Xen relocates the bottom 1MB at boot time.  You might find that
>>>> 32-bit Xen boots faster.
>>> 
>>> Could please expand a little on why the relocation causes it to slow down?
>> 
>> IIRC, the relocation itself is very slow because Xen just copies the low
>> 1MB, which includes the VGA hole, so there are a lot of emulated reads.
> 
> That seems a bit odd - ISA space reads can have side effects.

We should probably not do that then. :-)

In fact we only do it because the code was more convenient that way. The
first thing we do with out relocated copy of the bottom megabyte is zap it
with poison bytes.

 -- Keir

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



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

WARNING - OLD ARCHIVES

xen-devel

Re: [Xen-devel] Testing nested virtualization on Intel CPUs