From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-1.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_PASS autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 996CCC4360F for ; Mon, 1 Apr 2019 14:11:59 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 64992218FF for ; Mon, 1 Apr 2019 14:11:59 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727190AbfDAOL5 (ORCPT ); Mon, 1 Apr 2019 10:11:57 -0400 Received: from mx1.redhat.com ([209.132.183.28]:42404 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726542AbfDAOL5 (ORCPT ); Mon, 1 Apr 2019 10:11:57 -0400 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 5FB7230832F4; Mon, 1 Apr 2019 14:11:56 +0000 (UTC) Received: from [10.36.118.81] (unknown [10.36.118.81]) by smtp.corp.redhat.com (Postfix) with ESMTP id DC96A5D70E; Mon, 1 Apr 2019 14:11:42 +0000 (UTC) Subject: Re: On guest free page hinting and OOM From: David Hildenbrand To: "Michael S. Tsirkin" Cc: Nitesh Narayan Lal , kvm@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, pbonzini@redhat.com, lcapitulino@redhat.com, pagupta@redhat.com, wei.w.wang@intel.com, yang.zhang.wz@gmail.com, riel@surriel.com, dodgen@google.com, konrad.wilk@oracle.com, dhildenb@redhat.com, aarcange@redhat.com, alexander.duyck@gmail.com References: <20190329084058-mutt-send-email-mst@kernel.org> <20190329104311-mutt-send-email-mst@kernel.org> <7a3baa90-5963-e6e2-c862-9cd9cc1b5f60@redhat.com> <20190329125034-mutt-send-email-mst@kernel.org> <20190401073007-mutt-send-email-mst@kernel.org> <29e11829-c9ac-a21b-b2f1-ed833e4ca449@redhat.com> Openpgp: preference=signencrypt Autocrypt: addr=david@redhat.com; prefer-encrypt=mutual; keydata= xsFNBFXLn5EBEAC+zYvAFJxCBY9Tr1xZgcESmxVNI/0ffzE/ZQOiHJl6mGkmA1R7/uUpiCjJ dBrn+lhhOYjjNefFQou6478faXE6o2AhmebqT4KiQoUQFV4R7y1KMEKoSyy8hQaK1umALTdL QZLQMzNE74ap+GDK0wnacPQFpcG1AE9RMq3aeErY5tujekBS32jfC/7AnH7I0v1v1TbbK3Gp XNeiN4QroO+5qaSr0ID2sz5jtBLRb15RMre27E1ImpaIv2Jw8NJgW0k/D1RyKCwaTsgRdwuK Kx/Y91XuSBdz0uOyU/S8kM1+ag0wvsGlpBVxRR/xw/E8M7TEwuCZQArqqTCmkG6HGcXFT0V9 PXFNNgV5jXMQRwU0O/ztJIQqsE5LsUomE//bLwzj9IVsaQpKDqW6TAPjcdBDPLHvriq7kGjt WhVhdl0qEYB8lkBEU7V2Yb+SYhmhpDrti9Fq1EsmhiHSkxJcGREoMK/63r9WLZYI3+4W2rAc UucZa4OT27U5ZISjNg3Ev0rxU5UH2/pT4wJCfxwocmqaRr6UYmrtZmND89X0KigoFD/XSeVv jwBRNjPAubK9/k5NoRrYqztM9W6sJqrH8+UWZ1Idd/DdmogJh0gNC0+N42Za9yBRURfIdKSb B3JfpUqcWwE7vUaYrHG1nw54pLUoPG6sAA7Mehl3nd4pZUALHwARAQABzSREYXZpZCBIaWxk ZW5icmFuZCA8ZGF2aWRAcmVkaGF0LmNvbT7CwX4EEwECACgFAljj9eoCGwMFCQlmAYAGCwkI BwMCBhUIAgkKCwQWAgMBAh4BAheAAAoJEE3eEPcA/4Na5IIP/3T/FIQMxIfNzZshIq687qgG 8UbspuE/YSUDdv7r5szYTK6KPTlqN8NAcSfheywbuYD9A4ZeSBWD3/NAVUdrCaRP2IvFyELj xoMvfJccbq45BxzgEspg/bVahNbyuBpLBVjVWwRtFCUEXkyazksSv8pdTMAs9IucChvFmmq3 jJ2vlaz9lYt/lxN246fIVceckPMiUveimngvXZw21VOAhfQ+/sofXF8JCFv2mFcBDoa7eYob s0FLpmqFaeNRHAlzMWgSsP80qx5nWWEvRLdKWi533N2vC/EyunN3HcBwVrXH4hxRBMco3jvM m8VKLKao9wKj82qSivUnkPIwsAGNPdFoPbgghCQiBjBe6A75Z2xHFrzo7t1jg7nQfIyNC7ez MZBJ59sqA9EDMEJPlLNIeJmqslXPjmMFnE7Mby/+335WJYDulsRybN+W5rLT5aMvhC6x6POK z55fMNKrMASCzBJum2Fwjf/VnuGRYkhKCqqZ8gJ3OvmR50tInDV2jZ1DQgc3i550T5JDpToh dPBxZocIhzg+MBSRDXcJmHOx/7nQm3iQ6iLuwmXsRC6f5FbFefk9EjuTKcLMvBsEx+2DEx0E UnmJ4hVg7u1PQ+2Oy+Lh/opK/BDiqlQ8Pz2jiXv5xkECvr/3Sv59hlOCZMOaiLTTjtOIU7Tq 7ut6OL64oAq+zsFNBFXLn5EBEADn1959INH2cwYJv0tsxf5MUCghCj/CA/lc/LMthqQ773ga uB9mN+F1rE9cyyXb6jyOGn+GUjMbnq1o121Vm0+neKHUCBtHyseBfDXHA6m4B3mUTWo13nid 0e4AM71r0DS8+KYh6zvweLX/LL5kQS9GQeT+QNroXcC1NzWbitts6TZ+IrPOwT1hfB4WNC+X 2n4AzDqp3+ILiVST2DT4VBc11Gz6jijpC/KI5Al8ZDhRwG47LUiuQmt3yqrmN63V9wzaPhC+ xbwIsNZlLUvuRnmBPkTJwwrFRZvwu5GPHNndBjVpAfaSTOfppyKBTccu2AXJXWAE1Xjh6GOC 8mlFjZwLxWFqdPHR1n2aPVgoiTLk34LR/bXO+e0GpzFXT7enwyvFFFyAS0Nk1q/7EChPcbRb hJqEBpRNZemxmg55zC3GLvgLKd5A09MOM2BrMea+l0FUR+PuTenh2YmnmLRTro6eZ/qYwWkC u8FFIw4pT0OUDMyLgi+GI1aMpVogTZJ70FgV0pUAlpmrzk/bLbRkF3TwgucpyPtcpmQtTkWS gDS50QG9DR/1As3LLLcNkwJBZzBG6PWbvcOyrwMQUF1nl4SSPV0LLH63+BrrHasfJzxKXzqg rW28CTAE2x8qi7e/6M/+XXhrsMYG+uaViM7n2je3qKe7ofum3s4vq7oFCPsOgwARAQABwsFl BBgBAgAPBQJVy5+RAhsMBQkJZgGAAAoJEE3eEPcA/4NagOsP/jPoIBb/iXVbM+fmSHOjEshl KMwEl/m5iLj3iHnHPVLBUWrXPdS7iQijJA/VLxjnFknhaS60hkUNWexDMxVVP/6lbOrs4bDZ NEWDMktAeqJaFtxackPszlcpRVkAs6Msn9tu8hlvB517pyUgvuD7ZS9gGOMmYwFQDyytpepo YApVV00P0u3AaE0Cj/o71STqGJKZxcVhPaZ+LR+UCBZOyKfEyq+ZN311VpOJZ1IvTExf+S/5 lqnciDtbO3I4Wq0ArLX1gs1q1XlXLaVaA3yVqeC8E7kOchDNinD3hJS4OX0e1gdsx/e6COvy qNg5aL5n0Kl4fcVqM0LdIhsubVs4eiNCa5XMSYpXmVi3HAuFyg9dN+x8thSwI836FoMASwOl C7tHsTjnSGufB+D7F7ZBT61BffNBBIm1KdMxcxqLUVXpBQHHlGkbwI+3Ye+nE6HmZH7IwLwV W+Ajl7oYF+jeKaH4DZFtgLYGLtZ1LDwKPjX7VAsa4Yx7S5+EBAaZGxK510MjIx6SGrZWBrrV TEvdV00F2MnQoeXKzD7O4WFbL55hhyGgfWTHwZ457iN9SgYi1JLPqWkZB0JRXIEtjd4JEQcx +8Umfre0Xt4713VxMygW0PnQt5aSQdMD58jHFxTk092mU+yIHj5LeYgvwSgZN4airXk5yRXl SE+xAvmumFBY Organization: Red Hat GmbH Message-ID: Date: Mon, 1 Apr 2019 16:11:42 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.4.0 MIME-Version: 1.0 In-Reply-To: <29e11829-c9ac-a21b-b2f1-ed833e4ca449@redhat.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.79 on 10.5.11.15 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.44]); Mon, 01 Apr 2019 14:11:56 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01.04.19 16:09, David Hildenbrand wrote: >>> Thinking about your approach, there is one elementary thing to notice: >>> >>> Giving the guest pages from the buffer while hinting requests are being >>> processed means that the guest can and will temporarily make use of more >>> memory than desired. Essentially up to the point where MADV_FREE is >>> finally called for the hinted pages. >> >> Right - but that seems like exactly the reverse of the issue with the current >> approach which is guest can temporarily use less memory than desired. >> >>> Even then the guest will logicall >>> make use of more memory than desired until core MM takes pages away. >> >> That sounds more like a host issue though. If it wants to >> it can use e.g. MAD_DONTNEED. > > Indeed. But MADV_DONTNEED is somewhat undesired for performance reasons. > You want to do the work when swapping not when hinting. > > But what I wanted to say here: Looking at the pure size of your guest > will at least not help you to identify if more memory than desired will > be used. > >> >>> So: >>> 1) Unmodified guests will make use of more memory than desired. >> >> One interesting possibility for this is to add the buffer memory >> by hotplug after the feature has been negotiated. >> I agree this sounds complex. > > Yes it is, and it goes into the direction of virtio-mem that essentially > does that. But bad news: memory hotplug is complicated stuff, both on > the hypervisor and guest side. And things like NUMA make it more involved. > > But even then, malicious guest can simply fake feature negotiation and > make use of all hotplugged memory. Won't work, at least not for > malicious guests. > >> >> But I have an idea: how about we include the hint size in the >> num_pages counter? Then unmodified guests put >> it in the balloon and don't use it. Modified ones >> will know to use it just for hinting. > > These are the nightmares I was talking about. I would like to decouple > this feature as far as possible from balloon inflation/deflation. > Ballooning is 4k based and might have other undesirable side effect. > Just because somebody wants to use page hinting does not mean he wants > to use ballooning. Effectively, many people will want to avoid > ballooning completely by using page hinting for their use case. > >> >> >>> 2) Malicious guests will make use of more memory than desired. >> >> Well this limitation is fundamental to balloon right? > > Yep, it is the fundamental issue of ballooning. If memory is available > right from the boot, the system is free to do with it whatever it wants. > (one of the main things virtio-mem will do differently/better) > >> If host wants to add tracking of balloon memory, it >> can enforce the limits. So far no one bothered, >> but maybe with this feature we should start to do that. > > I think I already had endless rants about why this is not possible. > Ballooning as it is currently implemented by virtio-balloon is broken by > design. Period. You can and never will be able to distinguish unmodified > guests from malicious guests. Please don't design new approaches based > on broken design. > >> >>> 3) Sane, modified guests will make use of more memory than desired. >>> >>> Instead, we could make our life much easier by doing the following: >>> >>> 1) Introduce a parameter to cap the amount of memory concurrently hinted >>> similar like you suggested, just don't consider it a buffer value. >>> "-device virtio-balloon,hinting_size=1G". This gives us control over the >>> hinting proceess. >>> >>> hinting_size=0 (default) disables hinting >>> >>> The admin can tweak the number along with memory requirements of the >>> guest. We can make suggestions (e.g. calculate depending on #cores,#size >>> of memory, or simply "1GB") >> >> So if it's all up to the guest and for the benefit of the guest, and >> with no cost/benefit to the host, then why are we supplying this value >> from the host? > > See 3), the admin has to be aware of hinting behavior. > >> >>> 2) In the guest, track the size of hints in progress, cap at the >>> hinting_size. >>> >>> 3) Document hinting behavior >>> >>> "When hinting is enabled, memory up to hinting_size might temporarily be >>> removed from your guest in order to be hinted to the hypervisor. This is >>> only for a very short time, but might affect applications. Consider the >>> hinting_size when sizing your guest. If your application was tested with >>> XGB and a hinting size of 1G is used, please configure X+1GB for the >>> guest. Otherwise, performance degradation might be possible." >> >> OK, so let's start with this. Now let us assume that guest follows >> the advice. We thus know that 1GB is not needed for guest applications. >> So why do we want to allow applications to still use this extra memory? > > If the application does not need the 1GB, the 1GB will be hinted to the > hypervisor and are effectively only a buffer for the OOM scenario. > (ignoring page cache discussions for now). > > "So why do we want to allow applications to still use this extra memory" > is the EXACT same issue you have with your buffer approach. Any guest > can make use of the buffer and you won't be able to detect it. Very same > problem. Only in your approach, the guest might agree to play nicely by > not making use of the 1G you provided. Just as if the application does > not need/use the additional 1GB. > > The interesting thing is most probably: Will the hinting size usually be > reasonable small? At least I guess a guest with 4TB of RAM will not > suddenly get a hinting size of hundreds of GB. Most probably also only > something in the range of 1GB. But this is an interesting question to > look into. > > Also, if the admin does not care about performance implications when > already close to hinting, no need to add the additional 1Gb to the ram size. "close to OOM" is what I meant. -- Thanks, David / dhildenb