From mboxrd@z Thu Jan  1 00:00:00 1970
From: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Subject: Re: [PATCH] KVM: x86: Drop unused return value of kvm_mmu_remove_some_alloc_mmu_pages
Date: Tue, 06 Dec 2011 19:52:29 +0900
Message-ID: <4EDDF3ED.3090704@oss.ntt.co.jp>
References: <4ED90C5C.3080304@siemens.com> <4EDDE6CC.9060106@oss.ntt.co.jp> <4EDDF106.7080602@redhat.com> <4EDDF14F.70804@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Jan Kiszka <jan.kiszka@siemens.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	kvm <kvm@vger.kernel.org>,
	Takuya Yoshikawa <takuya.yoshikawa@gmail.com>
To: Avi Kivity <avi@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from serv2.oss.ntt.co.jp ([222.151.198.100]:41889 "EHLO
	serv2.oss.ntt.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S933092Ab1LFKvg (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 6 Dec 2011 05:51:36 -0500
In-Reply-To: <4EDDF14F.70804@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

(2011/12/06 19:41), Avi Kivity wrote:
> On 12/06/2011 12:40 PM, Avi Kivity wrote:
>>>
>>> Is there any need for mmu_shrink()?
>>
>> Without it a user can easily pin large amounts of kernel memory by
>> filling guest memory with page tables and shadowing them all.
>
> It would make an interesting test case, btw, to try to crash the kernel
> this way, and exercise the shrinker.
>

If nobody is working on this mmu_shrink() issues, I will do something.

But what I saw from the code is that the current zapper is effective
for the cases there are some unsync children.

So we may need to zap rather forcibly for tdp.

	Takuya