From mboxrd@z Thu Jan  1 00:00:00 1970
From: Carsten Otte <cotte@de.ibm.com>
Subject: Re: [patch 10/12] [PATCH] kvm-s390: storage key interface
Date: Fri, 09 Dec 2011 14:37:19 +0100
Message-ID: <4EE20F0F.5000706@de.ibm.com>
References: <20111209112326.454852716@de.ibm.com> <20111209112752.097882009@de.ibm.com>  <OFEDB7DC8E.0D8BE463-ONC1257961.004633DF-C1257961.0046C1E7@de.ibm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Avi Kivity <avi@redhat.com>, borntrae@linux.vnet.ibm.com,
	Constantin Werner <Constantin.Werner@de.ibm.com>,
	heicars2@linux.vnet.ibm.com, huckc@linux.vnet.ibm.com,
	Jens Freimann <JFREI@de.ibm.com>, KVM <kvm@vger.kernel.org>,
	mschwid2@linux.vnet.ibm.com, Marcelo Tossati <mtosatti@redhat.com>
To: Joachim von Buttlar <joachim_von_buttlar@de.ibm.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from e06smtp18.uk.ibm.com ([195.75.94.114]:59711 "EHLO
	e06smtp18.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753402Ab1LINiA (ORCPT <rfc822;kvm@vger.kernel.org>);
	Fri, 9 Dec 2011 08:38:00 -0500
Received: from /spool/local
	by e06smtp18.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted
	for <kvm@vger.kernel.org> from <cotte@de.ibm.com>;
	Fri, 9 Dec 2011 13:37:54 -0000
Received: from d06av04.portsmouth.uk.ibm.com (d06av04.portsmouth.uk.ibm.com [9.149.37.216])
	by d06nrmr1407.portsmouth.uk.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id pB9Dbjon2101456
	for <kvm@vger.kernel.org>; Fri, 9 Dec 2011 13:37:45 GMT
Received: from d06av04.portsmouth.uk.ibm.com (loopback [127.0.0.1])
	by d06av04.portsmouth.uk.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id pB9DbjoF009011
	for <kvm@vger.kernel.org>; Fri, 9 Dec 2011 06:37:45 -0700
In-Reply-To: <OFEDB7DC8E.0D8BE463-ONC1257961.004633DF-C1257961.0046C1E7@de.ibm.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On 09.12.2011 13:52, Joachim von Buttlar wrote:
> Shouldn't it be:               page_set_storage_key(pte_val(*ptep), skey |
> _PAGE_CHANGED, 1);
>
> +		 		 		 /* avoid race clobbering changed bit
> */
> +		 		 		 pte_val(*ptep) |= _PAGE_SWC;
No, the guest GR/GC bits get set to the value userspace wants down
below (this is set storage key after all), and for the host we turn on 
Martins _PAGE_SWC software bit in the pte to make sure we don't 
underindicate changed. As far as I can tell, this should be just fine.

> Typo:	 		 /* put acc+f plus guest referenced and changed into the
will fix.