From patchwork Fri Dec 30 21:52:51 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yu Zhao X-Patchwork-Id: 13084550 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 63E35C4332F for ; Fri, 30 Dec 2022 21:53:15 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235523AbiL3VxN (ORCPT ); Fri, 30 Dec 2022 16:53:13 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43744 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235468AbiL3VxD (ORCPT ); Fri, 30 Dec 2022 16:53:03 -0500 Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id D2B661CFE5 for ; Fri, 30 Dec 2022 13:52:59 -0800 (PST) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-47ede4426e1so124363217b3.7 for ; Fri, 30 Dec 2022 13:52:59 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:from:subject:mime-version:message-id:date:from:to:cc:subject :date:message-id:reply-to; bh=8TFtch4afv11+liPGrcYUQIUPF8E0jytiv2ZtV4VP60=; b=asPESHfjUZVqU8QJN1ZRjtdLbyZ3asXwvb7J7xZeFOEkZdLReFK5zvD6iEweEPLRNO uMNcn1qJJP1VCQjdCQWbtIgvNNGCbyKbKLAKi5actBzYN3l6drHtqt2ovdoS2SOwaASF NdVzXqbMnDicqkUCQSYzivZA5q90Qqk77ytGWSmT0yBS/Of60q41AJzk4B1uRTM68dOZ YrElqL40Nhf8OrFL1PCZfvDG9Jd5EOj2KQnz9vcwJ1pmGwDSvlEQCjJsfN7Nsot/OKAJ 3q3d2pFsTNsrAkhElMMca0lXrTpYZJCkp6JSjcheVNR3Sn91EDanrNymPKWy77Rd+67c wbBA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:from:subject:mime-version:message-id:date:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=8TFtch4afv11+liPGrcYUQIUPF8E0jytiv2ZtV4VP60=; b=FbG+8dGtRAXMRhmaDTizRLHBnHMi87f/GALLCMbx2lanIg06HRA/4gJ3HtOe8wudxy dvnvq2PTcqxsxbKJdJu0sO0VOP76alvYlmnsr6KCowoCK+l1tqC7VuCTy/8p0Qn+u29J dePcUsL3/cI69Pc6NI6KjDPR67TBgXuEFdpMngK/6ErpQeQMu7DI8dCfzbSqXP/lF6hO cEpk6thTax+xpDmpZ4eaftrFGDBtBB/UNkWyBznxvHs7fyVayZYDrA3KftTTZmQBqyK9 XEsVBc5G6p4wz2dDvil0NnZvTlK6n8BInVdr8yzLqeYavpxtIC52DqfEcyUZhnWo4Nsc fnVQ== X-Gm-Message-State: AFqh2kqQM75Q180G6fmSgwFmLQk1tv2FrQ7lLdHtjog9DOIvSi99qDWi N/tieJF91Waj/BgRRAQmZRycg2m1XDo= X-Google-Smtp-Source: AMrXdXtb9FEN+IrQw0fMFoCYU37iBqM6LqwOvQ0wWXeG+vFQ3r+4lVRzKnphHsXxu7vtPxqZ4AKu3MbENKY= X-Received: from yuzhao.bld.corp.google.com ([2620:15c:183:200:81fe:2008:27c1:d0cb]) (user=yuzhao job=sendgmr) by 2002:a05:690c:830:b0:483:a506:cb0e with SMTP id by16-20020a05690c083000b00483a506cb0emr2053093ywb.123.1672437179072; Fri, 30 Dec 2022 13:52:59 -0800 (PST) Date: Fri, 30 Dec 2022 14:52:51 -0700 Message-Id: <20221230215252.2628425-1-yuzhao@google.com> Mime-Version: 1.0 X-Mailer: git-send-email 2.39.0.314.g84b9a713c41-goog Subject: [PATCH mm-unstable v2 1/2] mm: add vma_has_recency() From: Yu Zhao To: Andrew Morton Cc: Alexander Viro , Andrea Righi , Johannes Weiner , Michael Larabel , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@google.com, Yu Zhao Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org This patch adds vma_has_recency() to indicate whether a VMA may exhibit temporal locality that the LRU algorithm relies on. This function returns false for VMAs marked by VM_SEQ_READ or VM_RAND_READ. While the former flag indicates linear access, i.e., a special case of spatial locality, both flags indicate a lack of temporal locality, i.e., the reuse of an area within a relatively small duration. "Recency" is chosen over "locality" to avoid confusion between temporal and spatial localities. Before this patch, the active/inactive LRU only ignored the accessed bit from VMAs marked by VM_SEQ_READ. After this patch, the active/inactive LRU and MGLRU share the same logic: they both ignore the accessed bit if vma_has_recency() returns false. For the active/inactive LRU, the following fio test showed a [6, 8]% increase in IOPS when randomly accessing mapped files under memory pressure. kb=$(awk '/MemTotal/ { print $2 }' /proc/meminfo) kb=$((kb - 8*1024*1024)) modprobe brd rd_nr=1 rd_size=$kb dd if=/dev/zero of=/dev/ram0 bs=1M mkfs.ext4 /dev/ram0 mount /dev/ram0 /mnt/ swapoff -a fio --name=test --directory=/mnt/ --ioengine=mmap --numjobs=8 \ --size=8G --rw=randrw --time_based --runtime=10m \ --group_reporting The discussion that led to this patch is here [1]. Additional test results are available in that thread. [1] https://lore.kernel.org/r/Y31s%2FK8T85jh05wH@google.com/ Signed-off-by: Yu Zhao --- include/linux/mm_inline.h | 8 ++++++++ mm/memory.c | 7 +++---- mm/rmap.c | 42 +++++++++++++++++---------------------- mm/vmscan.c | 5 ++++- 4 files changed, 33 insertions(+), 29 deletions(-) diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index d1c1f211a86f..fe5b8449e14a 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -595,4 +595,12 @@ pte_install_uffd_wp_if_needed(struct vm_area_struct *vma, unsigned long addr, #endif } +static inline bool vma_has_recency(struct vm_area_struct *vma) +{ + if (vma->vm_flags & (VM_SEQ_READ | VM_RAND_READ)) + return false; + + return true; +} + #endif diff --git a/mm/memory.c b/mm/memory.c index 4000e9f017e0..ee72badad847 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1402,8 +1402,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb, force_flush = 1; } } - if (pte_young(ptent) && - likely(!(vma->vm_flags & VM_SEQ_READ))) + if (pte_young(ptent) && likely(vma_has_recency(vma))) mark_page_accessed(page); } rss[mm_counter(page)]--; @@ -5148,8 +5147,8 @@ static inline void mm_account_fault(struct pt_regs *regs, #ifdef CONFIG_LRU_GEN static void lru_gen_enter_fault(struct vm_area_struct *vma) { - /* the LRU algorithm doesn't apply to sequential or random reads */ - current->in_lru_fault = !(vma->vm_flags & (VM_SEQ_READ | VM_RAND_READ)); + /* the LRU algorithm only applies to accesses with recency */ + current->in_lru_fault = vma_has_recency(vma); } static void lru_gen_exit_fault(void) diff --git a/mm/rmap.c b/mm/rmap.c index 8a24b90d9531..9abffdd63a6a 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -823,25 +823,14 @@ static bool folio_referenced_one(struct folio *folio, } if (pvmw.pte) { - if (lru_gen_enabled() && pte_young(*pvmw.pte) && - !(vma->vm_flags & (VM_SEQ_READ | VM_RAND_READ))) { + if (lru_gen_enabled() && pte_young(*pvmw.pte)) { lru_gen_look_around(&pvmw); referenced++; } if (ptep_clear_flush_young_notify(vma, address, - pvmw.pte)) { - /* - * Don't treat a reference through - * a sequentially read mapping as such. - * If the folio has been used in another mapping, - * we will catch it; if this other mapping is - * already gone, the unmap path will have set - * the referenced flag or activated the folio. - */ - if (likely(!(vma->vm_flags & VM_SEQ_READ))) - referenced++; - } + pvmw.pte)) + referenced++; } else if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) { if (pmdp_clear_flush_young_notify(vma, address, pvmw.pmd)) @@ -875,7 +864,20 @@ static bool invalid_folio_referenced_vma(struct vm_area_struct *vma, void *arg) struct folio_referenced_arg *pra = arg; struct mem_cgroup *memcg = pra->memcg; - if (!mm_match_cgroup(vma->vm_mm, memcg)) + /* + * Ignore references from this mapping if it has no recency. If the + * folio has been used in another mapping, we will catch it; if this + * other mapping is already gone, the unmap path will have set the + * referenced flag or activated the folio in zap_pte_range(). + */ + if (!vma_has_recency(vma)) + return true; + + /* + * If we are reclaiming on behalf of a cgroup, skip counting on behalf + * of references from different cgroups. + */ + if (memcg && !mm_match_cgroup(vma->vm_mm, memcg)) return true; return false; @@ -906,6 +908,7 @@ int folio_referenced(struct folio *folio, int is_locked, .arg = (void *)&pra, .anon_lock = folio_lock_anon_vma_read, .try_lock = true, + .invalid_vma = invalid_folio_referenced_vma, }; *vm_flags = 0; @@ -921,15 +924,6 @@ int folio_referenced(struct folio *folio, int is_locked, return 1; } - /* - * If we are reclaiming on behalf of a cgroup, skip - * counting on behalf of references from different - * cgroups - */ - if (memcg) { - rwc.invalid_vma = invalid_folio_referenced_vma; - } - rmap_walk(folio, &rwc); *vm_flags = pra.vm_flags; diff --git a/mm/vmscan.c b/mm/vmscan.c index 6929402db149..cdf96aec39dc 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -3794,7 +3794,10 @@ static int should_skip_vma(unsigned long start, unsigned long end, struct mm_wal if (is_vm_hugetlb_page(vma)) return true; - if (vma->vm_flags & (VM_LOCKED | VM_SPECIAL | VM_SEQ_READ | VM_RAND_READ)) + if (!vma_has_recency(vma)) + return true; + + if (vma->vm_flags & (VM_LOCKED | VM_SPECIAL)) return true; if (vma == get_gate_vma(vma->vm_mm)) From patchwork Fri Dec 30 21:52:52 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yu Zhao X-Patchwork-Id: 13084551 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 037D2C3DA7C for ; Fri, 30 Dec 2022 21:53:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235497AbiL3VxP (ORCPT ); Fri, 30 Dec 2022 16:53:15 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43840 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235487AbiL3VxL (ORCPT ); Fri, 30 Dec 2022 16:53:11 -0500 Received: from mail-yb1-xb49.google.com (mail-yb1-xb49.google.com [IPv6:2607:f8b0:4864:20::b49]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id DC2D21CFE8 for ; Fri, 30 Dec 2022 13:53:00 -0800 (PST) Received: by mail-yb1-xb49.google.com with SMTP id v7-20020a056902108700b007971863ae72so2149243ybu.22 for ; Fri, 30 Dec 2022 13:53:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:from:subject:references:mime-version:message-id:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=O4DokS/FZJlMuz65NYBFzd2Wq7AgsGKfPd6D2qH+qms=; b=FgGwqmDX00GIWfIYPn/v8jRlZd2/7/eGhsa3XhZaxfOffHYe3todQjnwAAsP2bTkVA NXtj5qc/YfA9UTTSKVXDsNyMAQiVWsVl5biYy5PbMY3paPPCX1TsCSWPWHZT0vluz8M2 VScUMyAHmNjsnk5aY8zH9ZZjn9mmLgCd7ClnRRUHlssSxX4w6uZ375mhpBYm4hH6bUqk xB7kwz3GbYZZCAfOS0HKy4R4y5kIVxhSP1t+NvJeST2lASf5xuxdPggulq/bzWk8UuQ1 hAI2KXL04UliLjieBCQZ/eRt4IbXiY4KhcH6y4MZ0aG+I4BPczrrrZjPSI8kHJD1RcBs 60eQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:from:subject:references:mime-version:message-id:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=O4DokS/FZJlMuz65NYBFzd2Wq7AgsGKfPd6D2qH+qms=; b=sVGYYWKxZc5UzO5I85W3RjVJ+IK87PbUmcvMc4tiQLcPdqO/EFmsuWUrAn/QQVe5UW 3pUqbB/7OPOwEV6wle8nQhBaV6Y/qofyCtSJ84OD8YgQjljN19VjU3i3D/O1qDhzlSew hrJu6FnYnFQsTC0eHIZQZGhI5sKizz/8JrmNSA+sgrUGIDmIgMR8fQRlQFYtngDiHxY+ I7rzRmkB5e0hHFRhyHpProk3QYPx+vHI6tURgIfHXAzZjmPt9dd3Qv4DJ4myAZSG/Ga3 qQcnauxNZ7AJIhKeCA5NDFH9GDr1awrmJvdRmrsr7jZvxYTDAVkoWQ3/Ev6SfCCI0bum Hy3Q== X-Gm-Message-State: AFqh2kpb1+yo17BnUAOZ1nWhR+Umu9gyJ9LO57CcVDOHh0SmXPdoVXum KgUEbysjY47P1zFm2M+FVRc64rlVxKo= X-Google-Smtp-Source: AMrXdXtTmFzViPQcW6Bc4aF7poiGpJjp1VELwrivuun/rdXIksgRPlpJqcuBPyGHvQEmV73KcdhsYhrdZTI= X-Received: from yuzhao.bld.corp.google.com ([2620:15c:183:200:81fe:2008:27c1:d0cb]) (user=yuzhao job=sendgmr) by 2002:a81:494f:0:b0:480:c531:5824 with SMTP id w76-20020a81494f000000b00480c5315824mr2177473ywa.247.1672437180183; Fri, 30 Dec 2022 13:53:00 -0800 (PST) Date: Fri, 30 Dec 2022 14:52:52 -0700 In-Reply-To: <20221230215252.2628425-1-yuzhao@google.com> Message-Id: <20221230215252.2628425-2-yuzhao@google.com> Mime-Version: 1.0 References: <20221230215252.2628425-1-yuzhao@google.com> X-Mailer: git-send-email 2.39.0.314.g84b9a713c41-goog Subject: [PATCH mm-unstable v2 2/2] mm: support POSIX_FADV_NOREUSE From: Yu Zhao To: Andrew Morton Cc: Alexander Viro , Andrea Righi , Johannes Weiner , Michael Larabel , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@google.com, Yu Zhao Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org This patch adds POSIX_FADV_NOREUSE to vma_has_recency() so that the LRU algorithm can ignore access to mapped files marked by this flag. The advantages of POSIX_FADV_NOREUSE are: 1. Unlike MADV_SEQUENTIAL and MADV_RANDOM, it does not alter the default readahead behavior. 2. Unlike MADV_SEQUENTIAL and MADV_RANDOM, it does not split VMAs and therefore does not take mmap_lock. 3. Unlike MADV_COLD, setting it has a negligible cost, regardless of how many pages it affects. Its limitations are: 1. Like POSIX_FADV_RANDOM and POSIX_FADV_SEQUENTIAL, it currently does not support range. IOW, its scope is the entire file. 2. It currently does not ignore access through file descriptors. Specifically, for the active/inactive LRU, given a file page shared by two users and one of them having set POSIX_FADV_NOREUSE on the file, this page will be activated upon the second user accessing it. This corner case can be covered by checking POSIX_FADV_NOREUSE before calling folio_mark_accessed() on the read path. But it is considered not worth the effort. There have been a few attempts to support POSIX_FADV_NOREUSE, e.g., [1]. This time the goal is to fill a niche: a few desktop applications, e.g., large file transferring and video encoding/decoding, want fast file streaming with mmap() rather than direct IO. Among those applications, an SVT-AV1 regression was reported when running with MGLRU [2]. The following test can reproduce that regression. kb=$(awk '/MemTotal/ { print $2 }' /proc/meminfo) kb=$((kb - 8*1024*1024)) modprobe brd rd_nr=1 rd_size=$kb dd if=/dev/zero of=/dev/ram0 bs=1M mkfs.ext4 /dev/ram0 mount /dev/ram0 /mnt/ swapoff -a fallocate -l 8G /mnt/swapfile mkswap /mnt/swapfile swapon /mnt/swapfile wget http://ultravideo.cs.tut.fi/video/Bosphorus_3840x2160_120fps_420_8bit_YUV_Y4M.7z 7z e -o/mnt/ Bosphorus_3840x2160_120fps_420_8bit_YUV_Y4M.7z SvtAv1EncApp --preset 12 -w 3840 -h 2160 \ -i /mnt/Bosphorus_3840x2160.y4m For MGLRU, the following change showed a [9-11]% increase in FPS, which makes it on par with the active/inactive LRU. patch Source/App/EncApp/EbAppMain.c < #include 35d35 < #include /* _O_BINARY */ 117a118 > posix_fadvise(config->mmap.fd, 0, 0, POSIX_FADV_NOREUSE); EOF [1] https://lore.kernel.org/r/1308923350-7932-1-git-send-email-andrea@betterlinux.com/ [2] https://openbenchmarking.org/result/2209259-PTS-MGLRU8GB57 Signed-off-by: Yu Zhao --- include/linux/fs.h | 2 ++ include/linux/mm_inline.h | 3 +++ mm/fadvise.c | 5 ++++- 3 files changed, 9 insertions(+), 1 deletion(-) diff --git a/include/linux/fs.h b/include/linux/fs.h index 066555ad1bf8..5660ed0edf1a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -166,6 +166,8 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset, /* File supports DIRECT IO */ #define FMODE_CAN_ODIRECT ((__force fmode_t)0x400000) +#define FMODE_NOREUSE ((__force fmode_t)0x800000) + /* File was opened by fanotify and shouldn't generate fanotify events */ #define FMODE_NONOTIFY ((__force fmode_t)0x4000000) diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index fe5b8449e14a..064f92c78bfa 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -600,6 +600,9 @@ static inline bool vma_has_recency(struct vm_area_struct *vma) if (vma->vm_flags & (VM_SEQ_READ | VM_RAND_READ)) return false; + if (vma->vm_file && (vma->vm_file->f_mode & FMODE_NOREUSE)) + return false; + return true; } diff --git a/mm/fadvise.c b/mm/fadvise.c index bf04fec87f35..fb7c5f43fd2a 100644 --- a/mm/fadvise.c +++ b/mm/fadvise.c @@ -80,7 +80,7 @@ int generic_fadvise(struct file *file, loff_t offset, loff_t len, int advice) case POSIX_FADV_NORMAL: file->f_ra.ra_pages = bdi->ra_pages; spin_lock(&file->f_lock); - file->f_mode &= ~FMODE_RANDOM; + file->f_mode &= ~(FMODE_RANDOM | FMODE_NOREUSE); spin_unlock(&file->f_lock); break; case POSIX_FADV_RANDOM: @@ -107,6 +107,9 @@ int generic_fadvise(struct file *file, loff_t offset, loff_t len, int advice) force_page_cache_readahead(mapping, file, start_index, nrpages); break; case POSIX_FADV_NOREUSE: + spin_lock(&file->f_lock); + file->f_mode |= FMODE_NOREUSE; + spin_unlock(&file->f_lock); break; case POSIX_FADV_DONTNEED: __filemap_fdatawrite_range(mapping, offset, endbyte,