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=-5.3 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,NICE_REPLY_A, SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=no 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 235A6C433DB for ; Tue, 9 Feb 2021 10:23:57 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 530A164E6F for ; Tue, 9 Feb 2021 10:23:56 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 530A164E6F Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=redhat.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 6AAC56B006E; Tue, 9 Feb 2021 05:23:55 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 658396B0070; Tue, 9 Feb 2021 05:23:55 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 51FD46B0071; Tue, 9 Feb 2021 05:23:55 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0210.hostedemail.com [216.40.44.210]) by kanga.kvack.org (Postfix) with ESMTP id 350F16B006E for ; Tue, 9 Feb 2021 05:23:55 -0500 (EST) Received: from smtpin15.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with ESMTP id F1BB812F9 for ; Tue, 9 Feb 2021 10:23:54 +0000 (UTC) X-FDA: 77798343588.15.jail77_2308b1027606 Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin15.hostedemail.com (Postfix) with ESMTP id C14E41814B0D0 for ; Tue, 9 Feb 2021 10:23:54 +0000 (UTC) X-HE-Tag: jail77_2308b1027606 X-Filterd-Recvd-Size: 10437 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [216.205.24.124]) by imf23.hostedemail.com (Postfix) with ESMTP for ; Tue, 9 Feb 2021 10:23:53 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1612866233; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=1IoX1DQCPa+nZtjLne30VDad7vc/+9XUN1LzFVohs0o=; b=CHLKqheCD8/67o9xr8llyKnonFmde21iE75kEqc1RUCbaiLFow1RgZCL8zXjBOnLqNc4gP 50t1gEPtrIKiTZj5CJDIBg1qizUzSi0SyOD9tuhRkJmz2TFCAz91n5TsJJJFFG73YIfHHJ dT/VfNWwbaxbIYgRVZjc3I6jl3tpass= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-65-shSTZr8_P1uvA3gtZUZXlw-1; Tue, 09 Feb 2021 05:23:49 -0500 X-MC-Unique: shSTZr8_P1uvA3gtZUZXlw-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 5220F804023; Tue, 9 Feb 2021 10:23:44 +0000 (UTC) Received: from [10.36.113.141] (ovpn-113-141.ams2.redhat.com [10.36.113.141]) by smtp.corp.redhat.com (Postfix) with ESMTP id 192F22CE01; Tue, 9 Feb 2021 10:23:35 +0000 (UTC) To: Michal Hocko Cc: Mike Rapoport , Andrew Morton , Alexander Viro , Andy Lutomirski , Arnd Bergmann , Borislav Petkov , Catalin Marinas , Christopher Lameter , Dan Williams , Dave Hansen , Elena Reshetova , "H. Peter Anvin" , Ingo Molnar , James Bottomley , "Kirill A. Shutemov" , Matthew Wilcox , Mark Rutland , Mike Rapoport , Michael Kerrisk , Palmer Dabbelt , Paul Walmsley , Peter Zijlstra , Rick Edgecombe , Roman Gushchin , Shakeel Butt , Shuah Khan , Thomas Gleixner , Tycho Andersen , Will Deacon , linux-api@vger.kernel.org, linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-fsdevel@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org, linux-nvdimm@lists.01.org, linux-riscv@lists.infradead.org, x86@kernel.org References: <20210208211326.GV242749@kernel.org> <1F6A73CF-158A-4261-AA6C-1F5C77F4F326@redhat.com> <662b5871-b461-0896-697f-5e903c23d7b9@redhat.com> From: David Hildenbrand Organization: Red Hat GmbH Subject: Re: [PATCH v17 00/10] mm: introduce memfd_secret system call to create "secret" memory areas Message-ID: Date: Tue, 9 Feb 2021 11:23:35 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.5.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 Content-Transfer-Encoding: quoted-printable X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: >>> A lot of unevictable memory is a concern regardless of CMA/ZONE_MOVAB= LE. >>> As I've said it is quite easy to land at the similar situation even w= ith >>> tmpfs/MAP_ANON|MAP_SHARED on swapless system. Neither of the two is >>> really uncommon. It would be even worse that those would be allowed t= o >>> consume both CMA/ZONE_MOVABLE. >> >> IIRC, tmpfs/MAP_ANON|MAP_SHARED memory >> a) Is movable, can land in ZONE_MOVABLE/CMA >> b) Can be limited by sizing tmpfs appropriately >> >> AFAIK, what you describe is a problem with memory overcommit, not with= zone >> imbalances (below). Or what am I missing? >=20 > It can be problem for both. If you have just too much of shm (do not > forget about MAP_SHARED|MAP_ANON which is much harder to size from an > admin POV) then migrateability doesn't really help because you need a > free memory to migrate. Without reclaimability this can easily become a > problem. That is why I am saying this is not really a new problem. > Swapless systems are not all that uncommon. I get your point, it's similar but still different. "no memory in the=20 system" vs. "plenty of unusable free memory available in the system". In many setups, memory for user space applications can go to=20 ZONE_MOVABLE just fine. ZONE_NORMAL etc. can be used for supporting user=20 space memory (e.g., page tables) and other kernel stuff. Like, have 4GB of ZONE_MOVABLE with 2GB of ZONE_NORMAL. Have an=20 application (database) that allocates 4GB of memory. Works just fine.=20 The zone ratio ends up being a problem for example with many processes=20 (-> many page tables). Not being able to put user space memory into the movable zone is a=20 special case. And we are introducing yet another special case here=20 (besides vfio, rdma, unmigratable huge pages like gigantic pages). With plenty of secretmem, looking at /proc/meminfo Total vs. Free can be=20 a big lie of how your system behaves. > =20 >>> One has to be very careful when relying on CMA or movable zones. This= is >>> definitely worth a comment in the kernel command line parameter >>> documentation. But this is not a new problem. >> >> I see the following thing worth documenting: >> >> Assume you have a system with 2GB of ZONE_NORMAL/ZONE_DMA and 4GB of >> ZONE_MOVABLE/CMA. >> >> Assume you make use of 1.5GB of secretmem. Your system might run into = OOM >> any time although you still have plenty of memory on ZONE_MOVAVLE (and= even >> swap!), simply because you are making excessive use of unmovable alloc= ations >> (for user space!) in an environment where you should not make excessiv= e use >> of unmovable allocations (e.g., where should page tables go?). >=20 > yes, you are right of course and I am not really disputing this. But I > would argue that 2:1 Movable/Normal is something to expect problems > already. "Lowmem" allocations can easily trigger OOM even without secre= t > mem in the picture. It all just takes to allocate a lot of GFP_KERNEL o= r > even GFP_{HIGH}USER. Really, it is CMA/MOVABLE that are elephant in the > room and one has to be really careful when relying on them. Right, it's all about what the setup actually needs. Sure, there are=20 cases where you need significantly more GFP_KERNEL/GFP_{HIGH}USER such=20 that a 2:1 ratio is not feasible. But I claim that these are corner cases= . Secretmem gives user space the option to allocate a lot of=20 GFP_{HIGH}USER memory. If I am not wrong, "ulimit -a" tells me that each=20 application on F33 can allocate 16 GiB (!) of secretmem. Which other ways do you know where random user space can do something=20 similar? I'd be curious what other scenarios there are where user space=20 can easily allocate a lot of unmovable memory. > =20 >> The existing controls (mlock limit) don't really match the current sem= antics >> of that memory. I repeat it once again: secretmem *currently* resemble= s >> long-term pinned memory, not mlocked memory. >=20 > Well, if we had a proper user space pinning accounting then I would > agree that there is a better model to use. But we don't. And previous > attempts to achieve that have failed. So I am afraid that we do not hav= e > much choice left than using mlock as a model. Yes, I agree. >=20 >> Things will change when >> implementing migration support for secretmem pages. Until then, the >> semantics are different and this should be spelled out. >> >> For long-term pinnings this is kind of obvious, still we're now docume= nting >> it because it's dangerous to not be aware of. Secretmem behaves exactl= y the >> same and I think this is worth spelling out: secretmem has the potenti= al of >> being used much more often than fairly special vfio/rdma/ ... >=20 > yeah I do agree that pinning is a problem for movable/CMA but most > people simply do not care about those. Movable is the thing for hoptlug > and a really weird fragmentation avoidance IIRC and CMA is mostly to + handling gigantic pages dynamically > handle crap HW. If those are to be used along with secret mem or > longterm GUP then they will constantly bump into corner cases. Do not > take me wrong, we should be looking at those problems, we should even > document them but I do not see this as anything new. We should probably > have a central place in Documentation explaining all those problems. I Exactly. > would be even happy to see an explicit note in the tunables - e.g. > configuring movable/normal in 2:1 will get you back to 32b times wrt. > low mem problems. In most setups, ratios of 1:1 up to 4:1 work reasonably well. Of course,=20 it's not applicable to all setups (obviously). For example, oVirt has been using ratios of 3:1 for a long time. (online=20 all memory to ZONE_MOVABLE in the guest, never hotplug more than 3x boot=20 memory size). Most distros end up onlining all hotplugged memory on bare=20 metal to ZONE_MOVABLE, and I've seen basically no bug reports related to=20 that. Highmem was a little different, yet similar. RHEL provided the bigmem=20 kernel with ratios of 60:4, which worked in many setups. The main=20 difference to highmem was that e.g., pagetables could be placed onto it.=20 So ratios like 18:1 are completely insane with ZONE_MOVABLE. I am constantly trying to fight for making more stuff MOVABLE instead of=20 going into the other direction (e.g., because it's easier to implement,=20 which feels like the wrong direction). Maybe I am the only person that really cares about ZONE_MOVABLE these=20 days :) I can't stop such new stuff from popping up, so at least I want=20 it to be documented. --=20 Thanks, David / dhildenb