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 Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id E627BCAC5A5 for ; Wed, 24 Sep 2025 15:52:23 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 4D70D8E0012; Wed, 24 Sep 2025 11:52:23 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 4AEAF8E000E; Wed, 24 Sep 2025 11:52:23 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 39CE98E0012; Wed, 24 Sep 2025 11:52:23 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0017.hostedemail.com [216.40.44.17]) by kanga.kvack.org (Postfix) with ESMTP id 26E2E8E000E for ; Wed, 24 Sep 2025 11:52:23 -0400 (EDT) Received: from smtpin25.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id D12671608F1 for ; Wed, 24 Sep 2025 15:52:22 +0000 (UTC) X-FDA: 83924585724.25.5F3D809 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by imf02.hostedemail.com (Postfix) with ESMTP id 7177C8000B for ; Wed, 24 Sep 2025 15:52:20 +0000 (UTC) Authentication-Results: imf02.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=b6Z91fIS; spf=pass (imf02.hostedemail.com: domain of david@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=david@redhat.com; dmarc=pass (policy=quarantine) header.from=redhat.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1758729140; h=from:from:sender: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:dkim-signature; bh=XzVixmsYq6hZpBtrPyWTw9f2SWf0I2tYU6xFVpkJp/M=; b=kUr4WvXGcoWYc/RB23j5utZEc4Pu/KyoZJfVASKbYLdoQHTBNoKAHJvJxW7FAGWfav1Rk/ Xn0v0qzpf0tvwKEXIEaDveFjA2cBMSs7Iqp8jBbpVhKKD6vGT4eWu1W7Exns5V8keacDL0 ez1K5OyvH+s5gJ5lGUOEkfmjM2dTmbg= ARC-Authentication-Results: i=1; imf02.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=b6Z91fIS; spf=pass (imf02.hostedemail.com: domain of david@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=david@redhat.com; dmarc=pass (policy=quarantine) header.from=redhat.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1758729140; a=rsa-sha256; cv=none; b=zTwjTS44I9s3wCnJu3sdxSwzRpAF4nZ4w64svojVy1zzp4F18sjMBn4AcUG3FgQBj4VAsb moSlDjN30w0BjN/SPCqT6+NnFyxIs7j+wG2DG9zjllNNJ0bUmfiPRKtQsfiKLKHv1o36B6 B1U9io06YMvQNejjtOTwGm6WjXtxK1I= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1758729139; 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:autocrypt:autocrypt; bh=XzVixmsYq6hZpBtrPyWTw9f2SWf0I2tYU6xFVpkJp/M=; b=b6Z91fISeIdURY45PqM4lMh+5hrPOsh1HpKBNT3oR5TGLOqFXQhTtQJRwnT5NLo6EVatDA P0awvTpk+J43tTZzUe4cWz01nxYNgaYTDtf090ks0aaCgeuq3xN+lM29U2H3lbxu0qzyvH uZRjJ2hLA4o0bFfe0hvr3RBfWRBp+bA= Received: from mail-wm1-f72.google.com (mail-wm1-f72.google.com [209.85.128.72]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-688-HhRBOpaXMheKh1lUQbyt1Q-1; Wed, 24 Sep 2025 11:52:18 -0400 X-MC-Unique: HhRBOpaXMheKh1lUQbyt1Q-1 X-Mimecast-MFC-AGG-ID: HhRBOpaXMheKh1lUQbyt1Q_1758729137 Received: by mail-wm1-f72.google.com with SMTP id 5b1f17b1804b1-46e1ac7110eso17519195e9.2 for ; Wed, 24 Sep 2025 08:52:18 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1758729137; x=1759333937; h=content-transfer-encoding:in-reply-to:autocrypt:content-language :from:references:cc:to:subject:user-agent:mime-version:date :message-id:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=XzVixmsYq6hZpBtrPyWTw9f2SWf0I2tYU6xFVpkJp/M=; b=aWk0kjk6RyinjhADwZh8r7EbugAWBts68qYaLFy6tKNq36qys2vFCVGllc7fS688m6 T8Aky6uAtxtQwLM49ey/ViAdCgleqlNaZ/p2cGa1umdE4Nb6GkSCj1CSpzIw7MNzx5Rc PfyF6jWhdJUT2i4/SobnWr6wuPsYPNzQog+iICshPNFPQtzE7+yBzgGT72JXiVlXMJp7 +rRcVHqejAU6NhVw1UNc8qcWSurxYGZtHIxM8y1fzl249dVMfglyRL0dihLUZTZx1BTg 5mjNRSAAsLkbi4rR/NdWgk7ppdYi6BBnyoY3WEjA68wVroO9CehdH021mkFgg/7A79m8 8CVw== X-Forwarded-Encrypted: i=1; AJvYcCV1w0RrdrKOLxm4hN55bKRa3ZkhUj1fdoIQtaANpESl1l4e2dhVSfjIM57qjn8URZo6qX13jKhBcA==@kvack.org X-Gm-Message-State: AOJu0Yyoina8tXJ5ZO1tHE6aonml9BNcThjrhI/7PoS5MOiJZCJLKVOr pwknpnj+LlmxnUG5WVPq8QveaTIiK8dhfqrdzZXAqErAaiMQ06wO9QdKg1s0pHJCkMLmVWPhiA7 GUQLdRokaEf4JSb5JhsWddkib/nymYivCF+sOf5Pau3HOax69+Qe7 X-Gm-Gg: ASbGncvWFxyx02cw5qsv3GJPXtgWFVHTiJVeMg7p5umlTKwLK5BaeuyWv8IRPs9Ej2P O9Bk3albo3VEkoe12v/d3L4rKf//sLSHzV/SHcoylackFyFzmDp3Xy+EHB6nSXvmLh3uesvVKbn JqZpDi6LDXCnruTmk1k8KfjaeIE1TBxnZeX+Q3XsFYPFjDTh83iN85AwLARd6S0Zun6P+DOTNLR CaKH+3hqKlar88afGQ+FKpXuZ6BDOvHNtDwoWZ59QZoYLggrmo7kQvDL0bMfEOvBfEHyRCIVeSC gQhJmXqdFLULDJAMiJTmPc5OWh19e8Pha81AiBc+9T8FisrBizHhPVheiQUPqFucnaS3ENWBqs3 T93pohKXmgD6KxhAAA6lBEwqaN6R/FmHei8KcuJ6O6qO3KpU2Cti9gecWrq5FTt1B+w== X-Received: by 2002:a05:600c:c4ac:b0:45b:868e:7f7f with SMTP id 5b1f17b1804b1-46e329f7c36mr3653495e9.17.1758729137098; Wed, 24 Sep 2025 08:52:17 -0700 (PDT) X-Google-Smtp-Source: AGHT+IHtIjjK5EmtNWj4tkdf++2VKpJZBRqmaieXYSJIlJmYT2U4IUZmCf1YtUJmt1dosqc2FgOgoA== X-Received: by 2002:a05:600c:c4ac:b0:45b:868e:7f7f with SMTP id 5b1f17b1804b1-46e329f7c36mr3653175e9.17.1758729136675; Wed, 24 Sep 2025 08:52:16 -0700 (PDT) Received: from ?IPV6:2003:d8:2f14:2400:afc:9797:137c:a25b? (p200300d82f1424000afc9797137ca25b.dip0.t-ipconnect.de. [2003:d8:2f14:2400:afc:9797:137c:a25b]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-46e2a996da2sm45549245e9.5.2025.09.24.08.52.15 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Wed, 24 Sep 2025 08:52:16 -0700 (PDT) Message-ID: <9b05b974-7478-4c99-9c4f-6593e0fd4f93@redhat.com> Date: Wed, 24 Sep 2025 17:52:14 +0200 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [DISCUSSION] Fixing bad pmd due to a race condition between change_prot_numa() and THP migration in pre-6.5 kernels. To: Harry Yoo Cc: Andrew Morton , Lorenzo Stoakes , Zi Yan , Baolin Wang , "Liam R . Howlett" , Nico Pache , Ryan Roberts , Dev Jain , Barry Song , Lance Yang , Kiryl Shutsemau , Hugh Dickins , Vlastimil Babka , Jann Horn , Pedro Falcato , Jane Chu , linux-mm@kvack.org, stable@vger.kernel.org References: <20250921232709.1608699-1-harry.yoo@oracle.com> <6e4f6a37-2449-4089-8b3d-234ba86878e2@redhat.com> From: David Hildenbrand Autocrypt: addr=david@redhat.com; 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 ZW5icmFuZCA8ZGF2aWRAcmVkaGF0LmNvbT7CwZoEEwEIAEQCGwMCF4ACGQEFCwkIBwICIgIG FQoJCAsCBBYCAwECHgcWIQQb2cqtc1xMOkYN/MpN3hD3AP+DWgUCaJzangUJJlgIpAAKCRBN 3hD3AP+DWhAxD/9wcL0A+2rtaAmutaKTfxhTP0b4AAp1r/eLxjrbfbCCmh4pqzBhmSX/4z11 opn2KqcOsueRF1t2ENLOWzQu3Roiny2HOU7DajqB4dm1BVMaXQya5ae2ghzlJN9SIoopTWlR 0Af3hPj5E2PYvQhlcqeoehKlBo9rROJv/rjmr2x0yOM8qeTroH/ZzNlCtJ56AsE6Tvl+r7cW 3x7/Jq5WvWeudKrhFh7/yQ7eRvHCjd9bBrZTlgAfiHmX9AnCCPRPpNGNedV9Yty2Jnxhfmbv Pw37LA/jef8zlCDyUh2KCU1xVEOWqg15o1RtTyGV1nXV2O/mfuQJud5vIgzBvHhypc3p6VZJ lEf8YmT+Ol5P7SfCs5/uGdWUYQEMqOlg6w9R4Pe8d+mk8KGvfE9/zTwGg0nRgKqlQXrWRERv cuEwQbridlPAoQHrFWtwpgYMXx2TaZ3sihcIPo9uU5eBs0rf4mOERY75SK+Ekayv2ucTfjxr Kf014py2aoRJHuvy85ee/zIyLmve5hngZTTe3Wg3TInT9UTFzTPhItam6dZ1xqdTGHZYGU0O otRHcwLGt470grdiob6PfVTXoHlBvkWRadMhSuG4RORCDpq89vu5QralFNIf3EysNohoFy2A LYg2/D53xbU/aa4DDzBb5b1Rkg/udO1gZocVQWrDh6I2K3+cCs7BTQRVy5+RARAA59fefSDR 9nMGCb9LbMX+TFAoIQo/wgP5XPyzLYakO+94GrgfZjfhdaxPXMsl2+o8jhp/hlIzG56taNdt VZtPp3ih1AgbR8rHgXw1xwOpuAd5lE1qNd54ndHuADO9a9A0vPimIes78Hi1/yy+ZEEvRkHk /kDa6F3AtTc1m4rbbOk2fiKzzsE9YXweFjQvl9p+AMw6qd/iC4lUk9g0+FQXNdRs+o4o6Qvy iOQJfGQ4UcBuOy1IrkJrd8qq5jet1fcM2j4QvsW8CLDWZS1L7kZ5gT5EycMKxUWb8LuRjxzZ 3QY1aQH2kkzn6acigU3HLtgFyV1gBNV44ehjgvJpRY2cC8VhanTx0dZ9mj1YKIky5N+C0f21 zvntBqcxV0+3p8MrxRRcgEtDZNav+xAoT3G0W4SahAaUTWXpsZoOecwtxi74CyneQNPTDjNg azHmvpdBVEfj7k3p4dmJp5i0U66Onmf6mMFpArvBRSMOKU9DlAzMi4IvhiNWjKVaIE2Se9BY FdKVAJaZq85P2y20ZBd08ILnKcj7XKZkLU5FkoA0udEBvQ0f9QLNyyy3DZMCQWcwRuj1m73D sq8DEFBdZ5eEkj1dCyx+t/ga6x2rHyc8Sl86oK1tvAkwBNsfKou3v+jP/l14a7DGBvrmlYjO 59o3t6inu6H7pt7OL6u6BQj7DoMAEQEAAcLBfAQYAQgAJgIbDBYhBBvZyq1zXEw6Rg38yk3e EPcA/4NaBQJonNqrBQkmWAihAAoJEE3eEPcA/4NaKtMQALAJ8PzprBEXbXcEXwDKQu+P/vts IfUb1UNMfMV76BicGa5NCZnJNQASDP/+bFg6O3gx5NbhHHPeaWz/VxlOmYHokHodOvtL0WCC 8A5PEP8tOk6029Z+J+xUcMrJClNVFpzVvOpb1lCbhjwAV465Hy+NUSbbUiRxdzNQtLtgZzOV Zw7jxUCs4UUZLQTCuBpFgb15bBxYZ/BL9MbzxPxvfUQIPbnzQMcqtpUs21CMK2PdfCh5c4gS sDci6D5/ZIBw94UQWmGpM/O1ilGXde2ZzzGYl64glmccD8e87OnEgKnH3FbnJnT4iJchtSvx yJNi1+t0+qDti4m88+/9IuPqCKb6Stl+s2dnLtJNrjXBGJtsQG/sRpqsJz5x1/2nPJSRMsx9 5YfqbdrJSOFXDzZ8/r82HgQEtUvlSXNaXCa95ez0UkOG7+bDm2b3s0XahBQeLVCH0mw3RAQg r7xDAYKIrAwfHHmMTnBQDPJwVqxJjVNr7yBic4yfzVWGCGNE4DnOW0vcIeoyhy9vnIa3w1uZ 3iyY2Nsd7JxfKu1PRhCGwXzRw5TlfEsoRI7V9A8isUCoqE2Dzh3FvYHVeX4Us+bRL/oqareJ CIFqgYMyvHj7Q06kTKmauOe4Nf0l0qEkIuIzfoLJ3qr5UyXc2hLtWyT9Ir+lYlX9efqh7mOY qIws/H2t In-Reply-To: X-Mimecast-Spam-Score: 0 X-Mimecast-MFC-PROC-ID: xDT8BT6zXUMjsnsPQYzIdCFF6fxd-tSHfDMQGUxmulM_1758729137 X-Mimecast-Originator: redhat.com Content-Language: en-US Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-Rspamd-Queue-Id: 7177C8000B X-Rspam-User: X-Rspamd-Server: rspam07 X-Stat-Signature: g9i5k7r4w5d6fdqphmxrupdmujn414wd X-HE-Tag: 1758729140-885580 X-HE-Meta: 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 +YSvUOh4 HBhEA/utgHQSCX8bFk0JUHw16J5FjF9fsUyDnROV2PQXZ2tudgBio79im0ber4Wc8Cd1xyQCLnc84TQMs6KFnVHd3CU5czFSbqwmUgiAbpwxZcwmGqknPnCOAJDke/R8H61mpFHIhtrOMdMQF49qDX1FRL30GWXq1YeS26Cm3K4XdxQXhI0uNBtWD6pdKcXQLuvY9+31ZUKeYs5Q9QKfF6qgk5XbclXruLaar8OE8/oZkYCJQ5q+A7cZum4t+xz1xPO5gSF22QgdFWbl5AfHwtYgIY2K4iJHQNs+AvrhVeOcMqAXmotgF34gvxnrZdCCPGtOFwT/h1vM/NOfA2+fpZMZgxW4X5YKUOdCNnUkLi5iMZOqRL1rOEhJmoGdkjlcJPMhftGtlMdAqdftyLMZjrri24yR5GTu8n0UGCxW1Nrh8j99ohGO8MiQdMPPRrdxITaSkwBh84Pv6I63z7WJJohdKDoiMp55JmzUa 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: List-Subscribe: List-Unsubscribe: On 24.09.25 13:54, Harry Yoo wrote: > On Tue, Sep 23, 2025 at 04:09:06PM +0200, David Hildenbrand wrote: >> On 23.09.25 13:46, Harry Yoo wrote: >>> On Tue, Sep 23, 2025 at 11:00:57AM +0200, David Hildenbrand wrote: >>>> On 22.09.25 01:27, Harry Yoo wrote: >>>> This is all because we are trying to be smart and walking page tables >>>> without the page table lock held. This is just absolutely nasty. >>> >>> commit 175ad4f1e7a2 ("mm: mprotect: use pmd_trans_unstable instead of >>> taking the pmd_lock") did this :( >> >> Right. I can understand why we would not want to grab the lock when there is >> a leaf page table. But everything else is just asking for trouble (as we saw >> :) ). >>>> What about the following check: >>>> >>>> if (is_swap_pmd(*pmd) || pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) { >>>> >>>> Couldn't we have a similar race there when we are concurrently migrating? >>> >>> An excellent point! I agree that there could be a similar race, >>> but with something other than "bad pmd" error. >> >> Right, instead we'd go into change_pte_range() where we check >> pmd_trans_unstable(). > > Uh, my brain hurts again... :) Yeah, that's why I gave up faster than you trying to understand these horrible, horrible functions in the stable kernel :) > > In case is_swap_pmd() or pmd_trans_huge() returned true, but another > kernel thread splits THP after we checked it, __split_huge_pmd() or > change_huge_pmd() will just return without actually splitting or changing > pmd entry, if it turns out that evaluating > (is_swap_pmd() || pmd_trans_huge() || pmd_devmap()) as true > was false positive due to race condition, because they both double check > after acquiring pmd lock: > > 1) __split_huge_pmd() checks if it's either pmd_trans_huge(), pmd_devmap() > or is_pmd_migration_entry() under pmd lock. > > 2) change_huge_pmd() checks if it's either is_swap_pmd(), > pmd_trans_huge(), or pmd_devmap() under pmd lock. > > And if either function simply returns because it was not a THP, > pmd migration entry, or pmd devmap, khugepaged cannot colleapse > huge page because we're holding mmap_lock in read mode. > > And then we call change_pte_range() and that's safe. > >> After that, I'm not sure ... maybe we'll just retry > > Or as you mentioned, if we are misled into thinking it is not a THP, > PMD devmap, or swap PMD due to race condition, we'd end up going into > change_pte_range(). > >> or we'll accidentally try treating it as a PTE table. > > But then pmd_trans_unstable() check should prevent us from treating > it as PTE table (and we're still holding mmap_lock here). > In such case we don't retry but skip it instead. > >> Looks like >> pmd_trans_unstable()->pud_none_or_trans_huge_or_dev_or_clear_bad() would > > I think you mean > pmd_trans_unstable()->pmd_none_or_trans_huge_or_clear_bad()? Yes! > >> return "0" >> in case we hit migration entry? :/ > > pmd_none_or_trans_huge_or_clear_bad() open-coded is_swap_pmd(), as it > eventually checks !pmd_none() && !pmd_present() case. Ah, right, I missed the pmd_present() while skimming over this extremely horrible function. So pmd_trans_unstable()->pmd_none_or_trans_huge_or_clear_bad() would return "1" and make us retry. > >>> It'd be more robust to do something like: >> >> That's also what I had in mind. But all this lockless stuff makes me a bit >> nervous :) > > Yeah the code is not very straightforward... :/ > > But technically the diff that I pasted here should be enough to fix > this... or do you have any alternative approach in mind? Hopefully, I'm not convinced this code is not buggy, but at least regarding concurrent migration it should be fine with that. -- Cheers David / dhildenb