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/
Home Products Support Community News


Re: [Xen-devel] RFC: [0/2] Remove netloop by lazy copying in netback

To: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Subject: Re: [Xen-devel] RFC: [0/2] Remove netloop by lazy copying in netback
From: Isaku Yamahata <yamahata@xxxxxxxxxxxxx>
Date: Thu, 29 Mar 2007 15:08:38 +0900
Cc: Xen Development Mailing List <xen-devel@xxxxxxxxxxxxxxxxxxx>, Keir Fraser <keir@xxxxxxxxxxxxx>
Delivery-date: Wed, 28 Mar 2007 23:07:41 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <20070328130516.GA6068@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/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <20070327111151.GA26126@xxxxxxxxxxxxxxxxxxx> <C22EDFAB.C4A6%keir@xxxxxxxxxxxxx> <20070327204640.GA30552@xxxxxxxxxxxxxxxxxxx> <20070328130516.GA6068@xxxxxxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/
On Wed, Mar 28, 2007 at 11:05:16PM +1000, Herbert Xu wrote:
> On Wed, Mar 28, 2007 at 06:46:40AM +1000, Herbert Xu wrote:
> > 
> > I've checked again and the accessed bit is certainly present on ppc
> > as well as ia64.  The only I don't know is if it's present on the
> > nested page tables on ia64 but I see no reasons why it wouldn't be.
> Here is a completely untested patch that demonstrates how this could
> be done on x86.

Xen/IA64 needs more than accessed bit for the optimized tlb flush
so that destroy_grant_host_mapping() flushes tlb and
it doesn't set any bit of domain_dirty_cpumask.
Can we make flush_tlb_mask() arch dependent and 
record something like GNTST_unuse in struct arch_vcpu?

Xen-devel mailing list