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]) by smtp.lore.kernel.org (Postfix) with ESMTP id BB546C0218D for ; Tue, 28 Jan 2025 09:47:23 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 44E1628021B; Tue, 28 Jan 2025 04:47:23 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 3D76D280216; Tue, 28 Jan 2025 04:47:23 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 229EF28021B; Tue, 28 Jan 2025 04:47:23 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 01D6F280216 for ; Tue, 28 Jan 2025 04:47:22 -0500 (EST) Received: from smtpin12.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id 7C2C3B11F1 for ; Tue, 28 Jan 2025 09:47:22 +0000 (UTC) X-FDA: 83056382724.12.F535E99 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by imf03.hostedemail.com (Postfix) with ESMTP id EBE4E2000F for ; Tue, 28 Jan 2025 09:47:19 +0000 (UTC) Authentication-Results: imf03.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=V9+vOfed; spf=pass (imf03.hostedemail.com: domain of david@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=david@redhat.com; dmarc=pass (policy=none) header.from=redhat.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1738057640; 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=WVOWkkrE+TIkSeiZWX/FWt30SOkhDf3aK6WbHWD9TC4=; b=CYK1PYDev9w/BHcOQAI3SWAeRFl/EDeuX4tYpHQt+jQ+LBKVFV6tgzn5MuoXeYT3kDeBxy P7HfeWPjHS5GqtvAiGrCHIkuz6miLicR6ynx76lvqVL3SwzeGy5p0POInQgNVLCvV2yOcY A7a+278uVgsU+5YYUC/yrK2JG3pZmaI= ARC-Authentication-Results: i=1; imf03.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=V9+vOfed; spf=pass (imf03.hostedemail.com: domain of david@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=david@redhat.com; dmarc=pass (policy=none) header.from=redhat.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1738057640; a=rsa-sha256; cv=none; b=o8LVmyQexjT/BfXfY10LU7BkiWAu5iYf+ks7jDc2V5GUslb5L2aKkCr2V6haEAyOtH9vkC J+wGDaBr4acRH39FZOwsmj91ZGMDJ/Rsvzu15nsNUXaEzDl4fnUwdEUvMSF94auhWt+4kk qRgY01jQ7DoaOPn372oEu3szWjB7Nqw= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1738057639; 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=WVOWkkrE+TIkSeiZWX/FWt30SOkhDf3aK6WbHWD9TC4=; b=V9+vOfed3qT7FdMGClemcnbEYZ+qh8kzjo9KLbxIigCxNBeiHx+crtTLKl5AAaLfOCm3J8 aAWwbEmvjRfUM1lC0Qho8Cge/htnTiJt9wrle/XD6fJuzRsDzhuK5vkqAYHnqDXHN26/pv Zh4JINCmDHcTFAkunGxc5Vq1E2UuWfM= 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-368-AeDJYHwFPmizrqT14k5ozA-1; Tue, 28 Jan 2025 04:47:17 -0500 X-MC-Unique: AeDJYHwFPmizrqT14k5ozA-1 X-Mimecast-MFC-AGG-ID: AeDJYHwFPmizrqT14k5ozA Received: by mail-wm1-f72.google.com with SMTP id 5b1f17b1804b1-4362552ce62so26391415e9.0 for ; Tue, 28 Jan 2025 01:47:17 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1738057636; x=1738662436; h=content-transfer-encoding:in-reply-to:organization: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=WVOWkkrE+TIkSeiZWX/FWt30SOkhDf3aK6WbHWD9TC4=; b=JA0uqV132tGWW51oOdeaqSGDZ2EFXUxDuVa4cZXrLOVFZnpaH//NeYOS8AnT/9bZnu LGDVAusvYz2X2Meud3M931ia1GDmLrZKszegLTOjlW9fz+Hl49NrinewkYP0qAbLd4uP lclIxt5rPO1pU3UPzaeoAEaJjbBMU75l8qF+CFnXfQlKyMhFKGUPZqvEXoF5dVKPkAMl aEg11vdqwvf+tPqg+cgFINnjRt5Yxsv8G+XhA3aEFyTWM/7Kh4sxot4QYsXCFMmxy7gb RkF6FyaYo5KbZbUsxFUtwqLpXsdNNcRgimzk38bFIDtR+YV/bRjqETiqXnEWZtxziJaC S4Vg== X-Gm-Message-State: AOJu0YxDvahJNeY6TKHnLJju4Bi7qKVPg7l81sVMKyT8ubIXufPC5qKl RRHxhlauxaPxkOV3Cxt+tO1o/OAQJjF461V7b1sLFGCqTiLiYastvkylFDAt4cBGtuDz0kzn7Zt 3xkg7R6e82opXzfMRIS+Vc4x4PkgQOWqOxj3CPqd0D/GbCNSp X-Gm-Gg: ASbGncsZh+qO9pI6ZYfZb9y64tv3tXJz6vgeTNne84ZNce252OCCIbHeYTz3LBvYir/ uIEzT+3ye4b68J5VQrxbvVmNR1rFOyr5TqLRTYd8pm9KU7w88BDw+vrD5IoOSvv4mPMheHWgrxC 9yEASTnwdi1RUtKLrYIdDzYW9vqLft5sG2tzAqEH6AOLL4O6oNN6w4qZXBljl1dJyJhAdm/ero8 qmden8nIHaH8glowCO1i1NSLfgxGDV5hRCVb7Ml8OMmXN/vVJuoKBGNpgdu9mtvX6Ty2ALbGC3h Kuq7m9qOA/fK/azkcS02WN0i6vN5ALCRy2Y0WRrShvOV9na6s8QEd5s0rdEr78aUFyHhzCdIQW0 Q/i4r6iXQ9b3WOp+LUJVje5d1uooQpJTX X-Received: by 2002:a05:600c:46ca:b0:434:a367:2bd9 with SMTP id 5b1f17b1804b1-438913dfd7fmr470318735e9.14.1738057636175; Tue, 28 Jan 2025 01:47:16 -0800 (PST) X-Google-Smtp-Source: AGHT+IG+9k4APRw8xSMDaM5HWlZTt9Y6CtcxPipnQOKSFGJ6ZtbhcYT+5O5y29YAO3usm6mdVI18bQ== X-Received: by 2002:a05:600c:46ca:b0:434:a367:2bd9 with SMTP id 5b1f17b1804b1-438913dfd7fmr470318445e9.14.1738057635751; Tue, 28 Jan 2025 01:47:15 -0800 (PST) Received: from ?IPV6:2003:cb:c73f:ce00:1be7:6d7f:3cc3:563d? (p200300cbc73fce001be76d7f3cc3563d.dip0.t-ipconnect.de. [2003:cb:c73f:ce00:1be7:6d7f:3cc3:563d]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-438bd501721sm161002865e9.9.2025.01.28.01.47.13 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Tue, 28 Jan 2025 01:47:14 -0800 (PST) Message-ID: Date: Tue, 28 Jan 2025 10:47:12 +0100 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: Direct I/O performance problems with 1GB pages To: Christoph Hellwig , Matthew Wilcox Cc: linux-mm@kvack.org, linux-block@vger.kernel.org, Muchun Song , Jane Chu References: 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 ZW5icmFuZCA8ZGF2aWRAcmVkaGF0LmNvbT7CwZgEEwEIAEICGwMGCwkIBwMCBhUIAgkKCwQW AgMBAh4BAheAAhkBFiEEG9nKrXNcTDpGDfzKTd4Q9wD/g1oFAl8Ox4kFCRKpKXgACgkQTd4Q 9wD/g1oHcA//a6Tj7SBNjFNM1iNhWUo1lxAja0lpSodSnB2g4FCZ4R61SBR4l/psBL73xktp rDHrx4aSpwkRP6Epu6mLvhlfjmkRG4OynJ5HG1gfv7RJJfnUdUM1z5kdS8JBrOhMJS2c/gPf wv1TGRq2XdMPnfY2o0CxRqpcLkx4vBODvJGl2mQyJF/gPepdDfcT8/PY9BJ7FL6Hrq1gnAo4 3Iv9qV0JiT2wmZciNyYQhmA1V6dyTRiQ4YAc31zOo2IM+xisPzeSHgw3ONY/XhYvfZ9r7W1l pNQdc2G+o4Di9NPFHQQhDw3YTRR1opJaTlRDzxYxzU6ZnUUBghxt9cwUWTpfCktkMZiPSDGd KgQBjnweV2jw9UOTxjb4LXqDjmSNkjDdQUOU69jGMUXgihvo4zhYcMX8F5gWdRtMR7DzW/YE BgVcyxNkMIXoY1aYj6npHYiNQesQlqjU6azjbH70/SXKM5tNRplgW8TNprMDuntdvV9wNkFs 9TyM02V5aWxFfI42+aivc4KEw69SE9KXwC7FSf5wXzuTot97N9Phj/Z3+jx443jo2NR34XgF 89cct7wJMjOF7bBefo0fPPZQuIma0Zym71cP61OP/i11ahNye6HGKfxGCOcs5wW9kRQEk8P9 M/k2wt3mt/fCQnuP/mWutNPt95w9wSsUyATLmtNrwccz63XOwU0EVcufkQEQAOfX3n0g0fZz Bgm/S2zF/kxQKCEKP8ID+Vz8sy2GpDvveBq4H2Y34XWsT1zLJdvqPI4af4ZSMxuerWjXbVWb T6d4odQIG0fKx4F8NccDqbgHeZRNajXeeJ3R7gAzvWvQNLz4piHrO/B4tf8svmRBL0ZB5P5A 2uhdwLU3NZuK22zpNn4is87BPWF8HhY0L5fafgDMOqnf4guJVJPYNPhUFzXUbPqOKOkL8ojk CXxkOFHAbjstSK5Ca3fKquY3rdX3DNo+EL7FvAiw1mUtS+5GeYE+RMnDCsVFm/C7kY8c2d0G NWkB9pJM5+mnIoFNxy7YBcldYATVeOHoY4LyaUWNnAvFYWp08dHWfZo9WCiJMuTfgtH9tc75 7QanMVdPt6fDK8UUXIBLQ2TWr/sQKE9xtFuEmoQGlE1l6bGaDnnMLcYu+Asp3kDT0w4zYGsx 5r6XQVRH4+5N6eHZiaeYtFOujp5n+pjBaQK7wUUjDilPQ5QMzIuCL4YjVoylWiBNknvQWBXS lQCWmavOT9sttGQXdPCC5ynI+1ymZC1ORZKANLnRAb0NH/UCzcsstw2TAkFnMEbo9Zu9w7Kv AxBQXWeXhJI9XQssfrf4Gusdqx8nPEpfOqCtbbwJMATbHyqLt7/oz/5deGuwxgb65pWIzufa N7eop7uh+6bezi+rugUI+w6DABEBAAHCwXwEGAEIACYCGwwWIQQb2cqtc1xMOkYN/MpN3hD3 AP+DWgUCXw7HsgUJEqkpoQAKCRBN3hD3AP+DWrrpD/4qS3dyVRxDcDHIlmguXjC1Q5tZTwNB boaBTPHSy/Nksu0eY7x6HfQJ3xajVH32Ms6t1trDQmPx2iP5+7iDsb7OKAb5eOS8h+BEBDeq 3ecsQDv0fFJOA9ag5O3LLNk+3x3q7e0uo06XMaY7UHS341ozXUUI7wC7iKfoUTv03iO9El5f XpNMx/YrIMduZ2+nd9Di7o5+KIwlb2mAB9sTNHdMrXesX8eBL6T9b+MZJk+mZuPxKNVfEQMQ a5SxUEADIPQTPNvBewdeI80yeOCrN+Zzwy/Mrx9EPeu59Y5vSJOx/z6OUImD/GhX7Xvkt3kq Er5KTrJz3++B6SH9pum9PuoE/k+nntJkNMmQpR4MCBaV/J9gIOPGodDKnjdng+mXliF3Ptu6 3oxc2RCyGzTlxyMwuc2U5Q7KtUNTdDe8T0uE+9b8BLMVQDDfJjqY0VVqSUwImzTDLX9S4g/8 kC4HRcclk8hpyhY2jKGluZO0awwTIMgVEzmTyBphDg/Gx7dZU1Xf8HFuE+UZ5UDHDTnwgv7E th6RC9+WrhDNspZ9fJjKWRbveQgUFCpe1sa77LAw+XFrKmBHXp9ZVIe90RMe2tRL06BGiRZr jPrnvUsUUsjRoRNJjKKA/REq+sAnhkNPPZ/NNMjaZ5b8Tovi8C0tmxiCHaQYqj7G2rgnT0kt WNyWQQ== Organization: Red Hat In-Reply-To: X-Mimecast-Spam-Score: 0 X-Mimecast-MFC-PROC-ID: -iqCYhKGfKQM_ZRU_x-a-8uCAivKx6UUNThzRJuATSE_1738057636 X-Mimecast-Originator: redhat.com Content-Language: en-US Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-Rspam-User: X-Rspamd-Server: rspam09 X-Rspamd-Queue-Id: EBE4E2000F X-Stat-Signature: gx9brjk6ya1xncuaeh9z4h7iwqt3zx8m X-HE-Tag: 1738057639-803707 X-HE-Meta: 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 Y47/NNkt 2uj9kEoRU5amctti7dz5joaWleRsV/BIMb/6wsXPS2hJxZvYnNANJP5D/GHEx0esP7gKtlwAWKFFE5ljZvxA1UpSZKjs+AyCuk5LUcWfaBq7SYIOqdQdYlGjvKfTRMQfQJP7XprHGG5ESM/9rhqtsuysSw8HCfh8I86nZkKKD8wtgOPziB2S/6CyTKZRgU3C0w2eX3ozMkk05z0Y4Dzm8wJs0Pz+VPkCyHVqdY+9FBEhEKkDmik5gnoF+Ofn9FknsAp/YnKsLgFWiUDuygNY8wN5pTDHNH5puYvhAWySK6HY0+7CnsXgtSgKAJDaQ/gv9NbQ+9mRiiSI4cyzwqH0qWOHGP2XttR1LDNdJ2Qg2X4pTd6muRM7ia0nY4HVb8xENvoR7M0tr/tWt6Dw= 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 28.01.25 06:56, Christoph Hellwig wrote: > I read through this and unfortunately have nothing useful to contribute > to the actual lock sharding. Just two semi-related bits: > > On Sun, Jan 26, 2025 at 12:46:45AM +0000, Matthew Wilcox wrote: >> Postgres are experimenting with doing direct I/O to 1GB hugetlb pages. >> Andres has gathered some performance data showing significantly worse >> performance with 1GB pages compared to 2MB pages. I sent a patch >> recently which improves matters [1], but problems remain. >> >> The primary problem we've identified is contention of folio->_refcount >> with a strong secondary contention on folio->_pincount. This is coming >> from the call chain: >> >> iov_iter_extract_pages -> >> gup_fast_fallback -> >> try_grab_folio_fast > > Eww, gup_fast_fallback sent me down the wrong path, as it suggests that > is is the fallback slow path, but it's not. It got renamed from > internal_get_user_pages_fast in commit 23babe1934d7 ("mm/gup: > consistently name GUP-fast functions"). While old name wasn't all > that great the new one including fallback is just horrible. Can we > please fix this up? Yes, I did that renaming as part of that series after the name was suggested during review. I got confused myself reading this report. internal_get_user_pages_fast() -> gup_fast_fallback() Was certainly an improvement. Naming is hard, we want to express "try fast, but fallback to slow if it fails". Maybe "gup_fast_with_fallback" might be clearer, not sure. > > The other thing is that the whole GUP machinery takes a reference > per page fragment it touches and not just per folio fragment. Right, that's what calling code expects: hands out pages, one ref per page. > I'm not sure how fair access to the atomics is, but I suspect the > multiple calls to increment/decrement them per operation probably > don't make the lock contention any better. The "advanced" logic in __get_user_pages() makes sure that if you GUP multiple pages covered by a single PMD/PUD, that we will adjust the refcount only twice (once deep down in the page table walker, then in __get_user_pages()). There is certainly a lot of room for improvement in these page table walkers ... -- Cheers, David / dhildenb