From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [patch 2/3] KVM: MMU: skip global pgtables on sync due to cr3
	switch
Date: Fri, 31 Oct 2008 17:36:03 -0200
Message-ID: <20081031193603.GC21772@dmt.cnet>
References: <20081025223111.498934405@localhost.localdomain> <20081025223243.863908646@localhost.localdomain> <49045426.1080807@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: kvm@vger.kernel.org
To: Avi Kivity <avi@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx2.redhat.com ([66.187.237.31]:42310 "EHLO mx2.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751364AbYJaThX (ORCPT <rfc822;kvm@vger.kernel.org>);
	Fri, 31 Oct 2008 15:37:23 -0400
Received: from int-mx2.corp.redhat.com (int-mx2.corp.redhat.com [172.16.27.26])
	by mx2.redhat.com (8.13.8/8.13.8) with ESMTP id m9VJbMGJ025807
	for <kvm@vger.kernel.org>; Fri, 31 Oct 2008 15:37:22 -0400
Content-Disposition: inline
In-Reply-To: <49045426.1080807@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Sun, Oct 26, 2008 at 01:27:34PM +0200, Avi Kivity wrote:
> Marcelo Tosatti wrote:
>> Skip syncing global pages on cr3 switch (but not on cr4/cr0). This is
>> important for Linux 32-bit guests with PAE, where the kmap page is
>> marked as global.
>>
>>   
>
> Patch is good, but won't apply without the first.
>
>>  {
>>  	u64 spte;
>>  	int ret = 0;
>>  	u64 mt_mask = shadow_mt_mask;
>> +	struct kvm_mmu_page *sp = page_header(__pa(shadow_pte));
>> +
>> +	if (!global && sp->global) {
>> +		sp->global = 0;
>>   
>
> A slight deficiency in this approach is that a page can't transition  
> from !global to global.  I don't think this is frequent, so we don't  
> need to deal with it.

Yep.

> But a more accurate approach is to keep a count  
> of non-global present mappings, and to activate the global logic when  
> this count is nonzero.

I don't think the overhead it adds is significant for the added
complexity?