WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

[Xen-devel] Re: [GIT PULL] xen /proc/mtrr implementation

To: Andi Kleen <andi@xxxxxxxxxxxxxx>
Subject: [Xen-devel] Re: [GIT PULL] xen /proc/mtrr implementation
From: Jeremy Fitzhardinge <jeremy@xxxxxxxx>
Date: Wed, 20 May 2009 15:49:17 -0700
Cc: Xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>, the arch/x86 maintainers <x86@xxxxxxxxxx>, Linux Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>, Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx>, "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>, Ingo Molnar <mingo@xxxxxxx>, Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Delivery-date: Wed, 20 May 2009 15:50:12 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <20090520225235.GJ8684@xxxxxxxxxxxxxxxxxx>
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>
References: <4A0DCC11.10307@xxxxxxxx> <m1my9ex818.fsf@xxxxxxxxxxxxxxxxx> <4A0DFF78.6000501@xxxxxxxx> <20090515202250.0f1218ef@jbarnes-g45> <m1iqk1k708.fsf@xxxxxxxxxxxxxxxxx> <4A10EAC4.9070701@xxxxxxxx> <20090518085902.GE10687@xxxxxxx> <4A11A3F8.1010202@xxxxxxxx> <87tz3gnq91.fsf@xxxxxxxxxxxxxxxxx> <4A143251.90208@xxxxxxxx> <20090520225235.GJ8684@xxxxxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Thunderbird 2.0.0.21 (X11/20090320)
Andi Kleen wrote:
There were drivers at some point that set WC MTRRs on their
transfer rings, not mmio.

To avoid cache pollution effects?

Maybe these all switched to PAT now, maybe not

It's hard to say because it could be also done from user space.

How does userspace do the virtual to physical mapping? If the driver can't also deal with the extra conversion from physical to machine under Xen, then it won't have much success either way.

   J

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