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 F2811C10F04 for ; Tue, 5 Dec 2023 14:17:03 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 87A6A6B0098; Tue, 5 Dec 2023 09:17:03 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 829C46B009D; Tue, 5 Dec 2023 09:17:03 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 6D3866B009E; Tue, 5 Dec 2023 09:17:03 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0015.hostedemail.com [216.40.44.15]) by kanga.kvack.org (Postfix) with ESMTP id 59D846B0098 for ; Tue, 5 Dec 2023 09:17:03 -0500 (EST) Received: from smtpin13.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 26751160162 for ; Tue, 5 Dec 2023 14:17:03 +0000 (UTC) X-FDA: 81532966326.13.8FBA3AD Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by imf20.hostedemail.com (Postfix) with ESMTP id 5435D1C0019 for ; Tue, 5 Dec 2023 14:17:00 +0000 (UTC) Authentication-Results: imf20.hostedemail.com; dkim=none; dmarc=pass (policy=none) header.from=arm.com; spf=pass (imf20.hostedemail.com: domain of ryan.roberts@arm.com designates 217.140.110.172 as permitted sender) smtp.mailfrom=ryan.roberts@arm.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1701785820; 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; bh=Cht6ahCfrzLBhe+fF9nekAtEzRODT0snbYwuzGinUiY=; b=EwWu12VIIytvBhXuAcmdqXgivCz/myaMTJQw8P0P+ZtbnqCods46PWVtSvMRaq9up61CFf TKrSbod/vyKURC2oUYIxYF7n1O52o2wH9Oxi2FtEgBXG+FOZieLuwOr8AqA84yH+BnmHct 8ktuCcc6z8xF+w/bFEY7OgZkb8K6Y0Y= ARC-Authentication-Results: i=1; imf20.hostedemail.com; dkim=none; dmarc=pass (policy=none) header.from=arm.com; spf=pass (imf20.hostedemail.com: domain of ryan.roberts@arm.com designates 217.140.110.172 as permitted sender) smtp.mailfrom=ryan.roberts@arm.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1701785820; a=rsa-sha256; cv=none; b=ZLRNrkk1JzpwrDHc167CGb4pWtx7GVM4jK9YYrCSqow+O3HIv5/PSI4hryyyvY+H2TNlWZ ZrxdbPrR20PG3duU1swCj9W9cRjKqYSQZiTdXmutWTOEODSAz6AwTJLL4Y6F0qggLfBOK+ mm18kdl+LbZHRItDUqJgAgk3opbcBcs= Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id DF00A139F; Tue, 5 Dec 2023 06:17:45 -0800 (PST) Received: from [10.57.73.130] (unknown [10.57.73.130]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 87FDD3F6C4; Tue, 5 Dec 2023 06:16:55 -0800 (PST) Message-ID: <0b48135a-a44b-4b5a-a33b-abd3a3b47ff8@arm.com> Date: Tue, 5 Dec 2023 14:16:53 +0000 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v3 01/15] mm: Batch-copy PTE ranges during fork() Content-Language: en-GB To: David Hildenbrand , Catalin Marinas , Will Deacon , Ard Biesheuvel , Marc Zyngier , Oliver Upton , James Morse , Suzuki K Poulose , Zenghui Yu , Andrey Ryabinin , Alexander Potapenko , Andrey Konovalov , Dmitry Vyukov , Vincenzo Frascino , Andrew Morton , Anshuman Khandual , Matthew Wilcox , Yu Zhao , Mark Rutland , Kefeng Wang , John Hubbard , Zi Yan , Barry Song <21cnbao@gmail.com>, Alistair Popple , Yang Shi Cc: linux-arm-kernel@lists.infradead.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org References: <20231204105440.61448-1-ryan.roberts@arm.com> <20231204105440.61448-2-ryan.roberts@arm.com> <104de2d6-ecf9-4b0c-a982-5bd8e1aea758@redhat.com> <5b8b9f8c-8e9b-42a5-b8b2-9b96903f3ada@redhat.com> From: Ryan Roberts In-Reply-To: Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Rspam-User: X-Rspamd-Server: rspam12 X-Rspamd-Queue-Id: 5435D1C0019 X-Stat-Signature: ri8nb8tdtahetiwydzuddncyfxk8abmo X-HE-Tag: 1701785820-415454 X-HE-Meta: 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 6uo/Frlp dcDVPvTX+21ZXZv3OTNVz+VUOcaHg7jGjRi42BBhUm10BBg5A+a797nm+6T4RASm2C/jwhCq2d4O/L+Bkn5hjGkOr4HsZ5zIIDqvWZi0kzfDwGK7jNO4S4LSP+nrtqQj2gMYgiC7C1OphxFo/r3ha+vfGP/tfYn84QbG03N7Mm/Gxs2U7FlVVoWSgdpwjru5UR6VNNMa+q9b0le6BM6021PR53bIb5ZrXlrA/KoRjZPHCKe6P4PAvXtEQDM2Dsy1lPnUFURRpQnefovn6TRp/MRsCRn2mrDlNajEe46woYRMdbRKhpsnK5orc99P9aj/5NA0VHWp+C0LqeMw= 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 05/12/2023 12:04, David Hildenbrand wrote: > On 05.12.23 12:30, Ryan Roberts wrote: >> On 04/12/2023 17:27, David Hildenbrand wrote: >>>> >>>> With rmap batching from [1] -- rebased+changed on top of that -- we could turn >>>> that into an effective (untested): >>>> >>>>            if (page && folio_test_anon(folio)) { >>>> +               nr = folio_nr_pages_cont_mapped(folio, page, src_pte, addr, >>>> end, >>>> +                                               pte, enforce_uffd_wp, >>>> &nr_dirty, >>>> +                                               &nr_writable); >>>>                    /* >>>>                     * If this page may have been pinned by the parent process, >>>>                     * copy the page immediately for the child so that we'll >>>> always >>>>                     * guarantee the pinned page won't be randomly replaced >>>> in the >>>>                     * future. >>>>                     */ >>>> -               folio_get(folio); >>>> -               if (unlikely(folio_try_dup_anon_rmap_pte(folio, page, >>>> src_vma))) { >>>> +               folio_ref_add(folio, nr); >>>> +               if (unlikely(folio_try_dup_anon_rmap_ptes(folio, page, nr, >>>> src_vma))) { >>>>                            /* Page may be pinned, we have to copy. */ >>>> -                       folio_put(folio); >>>> -                       return copy_present_page(dst_vma, src_vma, dst_pte, >>>> src_pte, >>>> -                                                addr, rss, prealloc, page); >>>> +                       folio_ref_sub(folio, nr); >>>> +                       ret = copy_present_page(dst_vma, src_vma, dst_pte, >>>> +                                               src_pte, addr, rss, prealloc, >>>> +                                               page); >>>> +                       return ret == 0 ? 1 : ret; >>>>                    } >>>> -               rss[MM_ANONPAGES]++; >>>> +               rss[MM_ANONPAGES] += nr; >>>>            } else if (page) { >>>> -               folio_get(folio); >>>> -               folio_dup_file_rmap_pte(folio, page); >>>> -               rss[mm_counter_file(page)]++; >>>> +               nr = folio_nr_pages_cont_mapped(folio, page, src_pte, addr, >>>> end, >>>> +                                               pte, enforce_uffd_wp, >>>> &nr_dirty, >>>> +                                               &nr_writable); >>>> +               folio_ref_add(folio, nr); >>>> +               folio_dup_file_rmap_ptes(folio, page, nr); >>>> +               rss[mm_counter_file(page)] += nr; >>>>            } >>>> >>>> >>>> We'll have to test performance, but it could be that we want to specialize >>>> more on !folio_test_large(). That code is very performance-sensitive. >>>> >>>> >>>> [1] https://lkml.kernel.org/r/20231204142146.91437-1-david@redhat.com >>> >>> So, on top of [1] without rmap batching but with a slightly modified version of >> >> Can you clarify what you mean by "without rmap batching"? I thought [1] >> implicitly adds rmap batching? (e.g. folio_dup_file_rmap_ptes(), which you've >> added in the code snippet above). > > Not calling the batched variants but essentially doing what your code does (with > some minor improvements, like updating the rss counters only once). > > The snipped above is only linked below. I had the performance numbers for [1] > ready, so I gave it a test on top of that. > > To keep it simple, you might just benchmark w and w/o your patches. > >> >>> yours (that keeps the existing code structure as pointed out and e.g., updates >>> counter updates), running my fork() microbenchmark with a 1 GiB of memory: >>> >>> Compared to [1], with all order-0 pages it gets 13--14% _slower_ and with all >>> PTE-mapped THP (order-9) it gets ~29--30% _faster_. >> >> What test are you running - I'd like to reproduce if possible, since it sounds >> like I've got some work to do to remove the order-0 regression. > > Essentially just allocating 1 GiB of memory an measuring how long it takes to > call fork(). > > order-0 benchmarks: > > https://gitlab.com/davidhildenbrand/scratchspace/-/raw/main/order-0-benchmarks.c?ref_type=heads > > e.g.,: $ ./order-0-benchmarks fork 100 > > > pte-mapped-thp benchmarks: > > https://gitlab.com/davidhildenbrand/scratchspace/-/raw/main/pte-mapped-thp-benchmarks.c?ref_type=heads > > e.g.,: $ ./pte-mapped-thp-benchmarks fork 100 > > > Ideally, pin to one CPU and get stable performance numbers by disabling > SMT+turbo etc. This is great - thanks! I'll get to work... > >> >>> >>> So looks like we really want to have a completely seprate code path for >>> "!folio_test_large()" to keep that case as fast as possible. And "Likely" we >>> want to use "likely(!folio_test_large()". ;) >> >> Yuk, but fair enough. If I can repro the perf numbers, I'll have a go a >> reworking this. >> >> I think you're also implicitly suggesting that this change needs to depend on >> [1]? Which is a shame... > > Not necessarily. It certainly cleans up the code, but we can do that in any > order reasonable. > >> >> I guess I should also go through a similar exercise for patch 2 in this series. > > > Yes. There are "unmap" and "pte-dontneed" benchmarks contained in both files above. >