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 13D8CC4706C for ; Tue, 9 Jan 2024 21:57:24 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 7EFAA8D001C; Tue, 9 Jan 2024 16:57:23 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 778708D0017; Tue, 9 Jan 2024 16:57:23 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 557668D001C; Tue, 9 Jan 2024 16:57:23 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0010.hostedemail.com [216.40.44.10]) by kanga.kvack.org (Postfix) with ESMTP id 3D7868D0017 for ; Tue, 9 Jan 2024 16:57:23 -0500 (EST) Received: from smtpin27.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay05.hostedemail.com (Postfix) with ESMTP id E7465409A8 for ; Tue, 9 Jan 2024 21:57:22 +0000 (UTC) X-FDA: 81661134324.27.6F2580D Received: from mail-ej1-f42.google.com (mail-ej1-f42.google.com [209.85.218.42]) by imf16.hostedemail.com (Postfix) with ESMTP id E971B180021 for ; Tue, 9 Jan 2024 21:57:20 +0000 (UTC) Authentication-Results: imf16.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=FcADvQzf; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf16.hostedemail.com: domain of souravpanda@google.com designates 209.85.218.42 as permitted sender) smtp.mailfrom=souravpanda@google.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1704837441; a=rsa-sha256; cv=none; b=YrZYK0IgBBsIuqDnPAxx5W6qM7hC+aBkEbuXrphkILVGHEq+U6uxe7KzX3kJw/HzhUILWs HYRoJv0gZkndJM6ydS6EKcvN7tVvjWZfnBEMwadS16b9qsLdGLC9MmxkL8ap/iJYYzDa3r RFwfhECYUaPxr0DtLWZpXTWKju1r8n8= ARC-Authentication-Results: i=1; imf16.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=FcADvQzf; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf16.hostedemail.com: domain of souravpanda@google.com designates 209.85.218.42 as permitted sender) smtp.mailfrom=souravpanda@google.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1704837441; 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: in-reply-to:in-reply-to:references:references:dkim-signature; bh=PjOtQPMLdVWYh5eikodvxNMttVHcMIZULVosJA/A02k=; b=iVDMBAWIEBwrhrYAx2qCK0zmTX1ykk1qH3eCPCeiSKpW/JWQlrbYZpGxNPH7RYyF0pFEzy 9kQRBWcGk30Jf7+ThBB8mttTCagjRTQCFhGHz2Bc4QDQqExPPLrWoNNIMLrwefYQaUP8FB SYGvaL39UVJRB0eKj/EmRPjK8BEUsNo= Received: by mail-ej1-f42.google.com with SMTP id a640c23a62f3a-a28f66dc7ffso792283566b.0 for ; Tue, 09 Jan 2024 13:57:20 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1704837439; x=1705442239; darn=kvack.org; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=PjOtQPMLdVWYh5eikodvxNMttVHcMIZULVosJA/A02k=; b=FcADvQzfZ2knsjsTP7LpYvDKIiYsp5hr9KeT8GTAoc87mWcjQeUXjMmBUKp+qrmZ1u vGzZ6nbWodOGmEjIfdCaIh7rhVtKKwmm3ZEG7DQogs58vQRXGYRRnq0Kp06llofdE+GK Asa0xVzRxp6AOalLhEnW2raq0DDylP7HWOJDDvgH/+w/RbT9UWjn6G/Z0CNZGDA1uHkJ QMFvtb/wUPMOU3BOBACb4IOiAyb/E6zYUPEE5fzAA5kOW0g8kjuCGqtSC1DMg6Qi1hFI 8hI5/b/YhOREx9eJVLj/oAA/APQa+hD4QEDN6UPtKOQB6KoqetFqIklyzf6YuZKdr0z9 1ccw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1704837439; x=1705442239; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=PjOtQPMLdVWYh5eikodvxNMttVHcMIZULVosJA/A02k=; b=Zfm8JGX3EaSslktLsVNBMHw6X+cFK0lnNeb0eWV0yMTfF5Jm41oc4WhjFPccbeROue YNhPNIm++UmrBYGL9ciB/ShS6zokbI1lSk25dRDn6W1lr17PKnn1tpzx2k1pssfmrTJc TFfH2d+/9+DPrQa5jhbyadfIklbyiiapHBNx5tr53qreFD6D25Baq+IZ3Ex65H6NEHln bhz+pITWNmSlmSfJrH1s39wQWoHtdH/+cimDv/3WODDchLr0UK+r1mIglzAVtC/1k6qA HXqGDBXA2CpmFjzeTc6VGawOfrL4W5iIyq3UgIPB69pNYsXtiVo9Zuoc6eomz56iPfX3 8WoQ== X-Gm-Message-State: AOJu0Yx4+kA2Tvm6VXRHwn8KLxtN06mFgSGxZ2x5hvLrHP+wvj5pXigq 8WAIIF2tNVbkHh2QY/6oINEPKwEHopYKKIaS+WcLn8r/klpX X-Google-Smtp-Source: AGHT+IFC711zo3/+0yiJG9A+YDZkuDYKqvM6T+Uru8xeput5OWyVx+/rrtP3t/d2AGRj4xtYD/PDIfi/wOvucAXaNSA= X-Received: by 2002:a17:906:c06:b0:a23:5939:759e with SMTP id s6-20020a1709060c0600b00a235939759emr149407ejf.26.1704837439199; Tue, 09 Jan 2024 13:57:19 -0800 (PST) MIME-Version: 1.0 References: <20231205223118.3575485-1-souravpanda@google.com> <20231205223118.3575485-2-souravpanda@google.com> <9266a9f9-71a5-dd88-df7b-facc037aaaff@google.com> In-Reply-To: <9266a9f9-71a5-dd88-df7b-facc037aaaff@google.com> From: Sourav Panda Date: Tue, 9 Jan 2024 13:57:07 -0800 Message-ID: Subject: Re: [PATCH v6 1/1] mm: report per-page metadata information To: David Rientjes Cc: corbet@lwn.net, gregkh@linuxfoundation.org, rafael@kernel.org, Andrew Morton , mike.kravetz@oracle.com, muchun.song@linux.dev, rppt@kernel.org, david@redhat.com, rdunlap@infradead.org, chenlinxuan@uniontech.com, yang.yang29@zte.com.cn, tomas.mudrunka@gmail.com, bhelgaas@google.com, ivan@cloudflare.com, pasha.tatashin@soleen.com, yosryahmed@google.com, hannes@cmpxchg.org, shakeelb@google.com, kirill.shutemov@linux.intel.com, wangkefeng.wang@huawei.com, adobriyan@gmail.com, Vlastimil Babka , "Liam R. Howlett" , surenb@google.com, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-doc@vger.kernel.org, linux-mm@kvack.org, Matthew Wilcox , weixugc@google.com Content-Type: multipart/alternative; boundary="000000000000f716e7060e8a651e" X-Rspam-User: X-Rspamd-Server: rspam06 X-Rspamd-Queue-Id: E971B180021 X-Stat-Signature: icrkaius65ac5dojj3c45de5wexfg9ub X-HE-Tag: 1704837440-392665 X-HE-Meta: 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 uJzPfSPd H6+f0SIY3+7pF8VemYiuE6aw2bHMQY/Cho8WDzQKB0blAv3Yy5LEKlUVMI+W6TkV2ZQ0JMwHIV65aGp23IkPucnDSzIelZFqtdaXkTzm5rE+qfaD93TxKoH7CEOXifE5mjeu8InieBfmpd1gAwYdSybEqugoutPeiy2IMcBJqB9OVVewj1ljnJfzBqekQ5LWBJrbJ5ZlW+ZjPXeU1E6du+NPuGyyINcFdt0oV3aII2SUDsukMqJfzzalHs4z0mQzhP2r9i1zA87AE2SUNt2teOYU6eync149Lb0o3s9kEu0DTrZt0PMxurEQFoS6tqKNtOVfK7BgDRbiEq8cNmXqlcQHAMK8G9hHl5U/GG93xW1TMm6uH/sw+ZGrkzzZ1EZVnNz5XrdDIe0m/xXrCtnlfNhsk93j150CXsrGWM/XhQ4jGIew= 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: --000000000000f716e7060e8a651e Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, Jan 5, 2024 at 4:38=E2=80=AFPM David Rientjes = wrote: > On Fri, 5 Jan 2024, David Rientjes wrote: > > > On Tue, 5 Dec 2023, Sourav Panda wrote: > > > > > diff --git a/Documentation/filesystems/proc.rst > b/Documentation/filesystems/proc.rst > > > index 49ef12df631b..d5901d04e082 100644 > > > --- a/Documentation/filesystems/proc.rst > > > +++ b/Documentation/filesystems/proc.rst > > > @@ -993,6 +993,7 @@ Example output. You may not have all of these > fields. > > > AnonPages: 4654780 kB > > > Mapped: 266244 kB > > > Shmem: 9976 kB > > > + PageMetadata: 513419 kB > > > KReclaimable: 517708 kB > > > Slab: 660044 kB > > > SReclaimable: 517708 kB > > > @@ -1095,6 +1096,8 @@ Mapped > > > files which have been mmapped, such as libraries > > > Shmem > > > Total memory used by shared memory (shmem) and tmpfs > > > +PageMetadata > > > + Memory used for per-page metadata > > > KReclaimable > > > Kernel allocations that the kernel will attempt to > reclaim > > > under memory pressure. Includes SReclaimable (below), > and other > > > diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c > > > index 45af9a989d40..f141bb2a550d 100644 > > > --- a/fs/proc/meminfo.c > > > +++ b/fs/proc/meminfo.c > > > @@ -39,7 +39,9 @@ static int meminfo_proc_show(struct seq_file *m, > void *v) > > > long available; > > > unsigned long pages[NR_LRU_LISTS]; > > > unsigned long sreclaimable, sunreclaim; > > > + unsigned long nr_page_metadata; > > > > Initialize it here (if we actually need this variable)? > > Thanks David! I will make this change in v7. > > > > int lru; > > > + int nid; > > > > > > si_meminfo(&i); > > > si_swapinfo(&i); > > > @@ -57,6 +59,10 @@ static int meminfo_proc_show(struct seq_file *m, > void *v) > > > sreclaimable =3D global_node_page_state_pages(NR_SLAB_RECLAIMABLE= _B); > > > sunreclaim =3D global_node_page_state_pages(NR_SLAB_UNRECLAIMABLE= _B); > > > > > > + nr_page_metadata =3D 0; > > > + for_each_online_node(nid) > > > + nr_page_metadata +=3D node_page_state(NODE_DATA(nid), > NR_PAGE_METADATA); > > > > Is this intended to be different than > > global_node_page_state_pages(NR_PAGE_METADATA)? > > We also present the metrics in /proc/meminfo to export the value in a readable format where all the values are consistently in kB (unlike vmstat with varying units). > > If so, any hint as to why we want to discount page metadata on offline > > nodes? We can't make an inference that metadata is always allocated > > locally, memoryless nodes need things like struct page allocated on > nodes > > with memory. > > > > Sorry, meant a node with only ZONE_MOVABLE here so metadata can't be > allocated locally. When we have a ZONE_MOVABLE=E2=80=99s page's metadata on another node, it w= ould NOT be discounted. It would be accounted for on the other node. > > But would be very interested to learn why this subtlety exists to sum up > only online nodes. > > > So even if a memoryless node is offline, we'd still be including its > > metadata here with the current implementation. > > > > Or maybe I'm missing a subtlety here for why this is not already > > global_node_page_state_pages(). > > If I understood correctly, as long as the memory is not hotplugged, it won't have any page metadata and thus we would NOT include its metadata. The moment the memory is hotplugged or hot-unplugged, the metadata will be incremented or decremented, correspondingly. > > > > + > > > show_val_kb(m, "MemTotal: ", i.totalram); > > > show_val_kb(m, "MemFree: ", i.freeram); > > > show_val_kb(m, "MemAvailable: ", available); > > > @@ -104,6 +110,7 @@ static int meminfo_proc_show(struct seq_file *m, > void *v) > > > show_val_kb(m, "Mapped: ", > > > global_node_page_state(NR_FILE_MAPPED)); > > > show_val_kb(m, "Shmem: ", i.sharedram); > > > + show_val_kb(m, "PageMetadata: ", nr_page_metadata); > > > show_val_kb(m, "KReclaimable: ", sreclaimable + > > > global_node_page_state(NR_KERNEL_MISC_RECLAIMABLE)); > > > show_val_kb(m, "Slab: ", sreclaimable + sunreclaim); > > > diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h > > > index 3c25226beeed..ef176152be7c 100644 > > > --- a/include/linux/mmzone.h > > > +++ b/include/linux/mmzone.h > > > @@ -207,6 +207,10 @@ enum node_stat_item { > > > PGPROMOTE_SUCCESS, /* promote successfully */ > > > PGPROMOTE_CANDIDATE, /* candidate pages to promote */ > > > #endif > > > + NR_PAGE_METADATA, /* Page metadata size (struct page and > page_ext) > > > + * in pages > > > + */ > > > + NR_PAGE_METADATA_BOOT, /* NR_PAGE_METADATA for bootmem */ > > > > So if some vmemmap pages are freed, then MemTotal could be incremented > by > > a portion of NR_PAGE_METADATA_BOOT and then this stat is decremented? > Is > > the goal that the sum of MemTotal + SUM(nr_page_metadata_boot) is alway= s > > constant? > > Yep, this is correct. > > > > NR_VM_NODE_STAT_ITEMS > > > }; > > > > > > diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h > > > index fed855bae6d8..af096a881f03 100644 > > > --- a/include/linux/vmstat.h > > > +++ b/include/linux/vmstat.h > > > @@ -656,4 +656,8 @@ static inline void lruvec_stat_sub_folio(struct > folio *folio, > > > { > > > lruvec_stat_mod_folio(folio, idx, -folio_nr_pages(folio)); > > > } > > > + > > > +void __init mod_node_early_perpage_metadata(int nid, long delta); > > > +void __init store_early_perpage_metadata(void); > > > + > > > #endif /* _LINUX_VMSTAT_H */ > > > diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c > > > index 87818ee7f01d..5b10d8d2b471 100644 > > > --- a/mm/hugetlb_vmemmap.c > > > +++ b/mm/hugetlb_vmemmap.c > > > @@ -230,10 +230,14 @@ static int vmemmap_remap_range(unsigned long > start, unsigned long end, > > > */ > > > static inline void free_vmemmap_page(struct page *page) > > > { > > > - if (PageReserved(page)) > > > + if (PageReserved(page)) { > > > free_bootmem_page(page); > > > - else > > > + mod_node_page_state(page_pgdat(page), > NR_PAGE_METADATA_BOOT, > > > + -1); > > > + } else { > > > __free_page(page); > > > + mod_node_page_state(page_pgdat(page), NR_PAGE_METADATA, > -1); > > > + } > > > } > > > > > > /* Free a list of the vmemmap pages */ > > > @@ -389,6 +393,7 @@ static int vmemmap_remap_free(unsigned long start= , > unsigned long end, > > > copy_page(page_to_virt(walk.reuse_page), > > > (void *)walk.reuse_addr); > > > list_add(&walk.reuse_page->lru, vmemmap_pages); > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA, 1); > > > } > > > > > > /* > > > @@ -437,14 +442,20 @@ static int alloc_vmemmap_page_list(unsigned lon= g > start, unsigned long end, > > > unsigned long nr_pages =3D (end - start) >> PAGE_SHIFT; > > > int nid =3D page_to_nid((struct page *)start); > > > struct page *page, *next; > > > + int i; > > > > > > - while (nr_pages--) { > > > + for (i =3D 0; i < nr_pages; i++) { > > > page =3D alloc_pages_node(nid, gfp_mask, 0); > > > - if (!page) > > > + if (!page) { > > > + mod_node_page_state(NODE_DATA(nid), > NR_PAGE_METADATA, > > > + i); > > > goto out; > > > + } > > > list_add(&page->lru, list); > > > } > > > > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA, nr_pages); > > > + > > > return 0; > > > out: > > > list_for_each_entry_safe(page, next, list, lru) > > > diff --git a/mm/mm_init.c b/mm/mm_init.c > > > index 077bfe393b5e..38f8e1f454a0 100644 > > > --- a/mm/mm_init.c > > > +++ b/mm/mm_init.c > > > @@ -26,6 +26,7 @@ > > > #include > > > #include > > > #include > > > +#include > > > #include "internal.h" > > > #include "slab.h" > > > #include "shuffle.h" > > > @@ -1656,6 +1657,8 @@ static void __init alloc_node_mem_map(struct > pglist_data *pgdat) > > > panic("Failed to allocate %ld bytes for node %d > memory map\n", > > > size, pgdat->node_id); > > > pgdat->node_mem_map =3D map + offset; > > > + mod_node_early_perpage_metadata(pgdat->node_id, > > > + DIV_ROUND_UP(size, > PAGE_SIZE)); > > > } > > > pr_debug("%s: node %d, pgdat %08lx, node_mem_map %08lx\n", > > > __func__, pgdat->node_id, (unsigned > long)pgdat, > > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > > > index 733732e7e0ba..dd78017105b0 100644 > > > --- a/mm/page_alloc.c > > > +++ b/mm/page_alloc.c > > > @@ -5636,6 +5636,7 @@ void __init setup_per_cpu_pageset(void) > > > for_each_online_pgdat(pgdat) > > > pgdat->per_cpu_nodestats =3D > > > alloc_percpu(struct per_cpu_nodestat); > > > + store_early_perpage_metadata(); > > > } > > > > > > __meminit void zone_pcp_init(struct zone *zone) > > > diff --git a/mm/page_ext.c b/mm/page_ext.c > > > index 4548fcc66d74..4ca9f298f34e 100644 > > > --- a/mm/page_ext.c > > > +++ b/mm/page_ext.c > > > @@ -201,6 +201,8 @@ static int __init alloc_node_page_ext(int nid) > > > return -ENOMEM; > > > NODE_DATA(nid)->node_page_ext =3D base; > > > total_usage +=3D table_size; > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA_BOOT, > > > + DIV_ROUND_UP(table_size, PAGE_SIZE)); > > > return 0; > > > } > > > > > > @@ -255,12 +257,15 @@ static void *__meminit alloc_page_ext(size_t > size, int nid) > > > void *addr =3D NULL; > > > > > > addr =3D alloc_pages_exact_nid(nid, size, flags); > > > - if (addr) { > > > + if (addr) > > > kmemleak_alloc(addr, size, 1, flags); > > > - return addr; > > > - } > > > + else > > > + addr =3D vzalloc_node(size, nid); > > > > > > - addr =3D vzalloc_node(size, nid); > > > + if (addr) { > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA, > > > + DIV_ROUND_UP(size, PAGE_SIZE)); > > > + } > > > > > > return addr; > > > } > > > @@ -303,18 +308,27 @@ static int __meminit > init_section_page_ext(unsigned long pfn, int nid) > > > > > > static void free_page_ext(void *addr) > > > { > > > + size_t table_size; > > > + struct page *page; > > > + struct pglist_data *pgdat; > > > + > > > + table_size =3D page_ext_size * PAGES_PER_SECTION; > > > + > > > if (is_vmalloc_addr(addr)) { > > > + page =3D vmalloc_to_page(addr); > > > + pgdat =3D page_pgdat(page); > > > vfree(addr); > > > } else { > > > - struct page *page =3D virt_to_page(addr); > > > - size_t table_size; > > > - > > > - table_size =3D page_ext_size * PAGES_PER_SECTION; > > > - > > > + page =3D virt_to_page(addr); > > > + pgdat =3D page_pgdat(page); > > > BUG_ON(PageReserved(page)); > > > kmemleak_free(addr); > > > free_pages_exact(addr, table_size); > > > } > > > + > > > + mod_node_page_state(pgdat, NR_PAGE_METADATA, > > > + -1L * (DIV_ROUND_UP(table_size, PAGE_SIZE))); > > > + > > > } > > > > > > static void __free_page_ext(unsigned long pfn) > > > diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c > > > index a2cbe44c48e1..054b49539843 100644 > > > --- a/mm/sparse-vmemmap.c > > > +++ b/mm/sparse-vmemmap.c > > > @@ -469,5 +469,13 @@ struct page * __meminit > __populate_section_memmap(unsigned long pfn, > > > if (r < 0) > > > return NULL; > > > > > > + if (system_state =3D=3D SYSTEM_BOOTING) { > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA_BOOT= , > > > + DIV_ROUND_UP(end - start, PAGE_SIZE))= ; > > > + } else { > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA, > > > + DIV_ROUND_UP(end - start, PAGE_SIZE))= ; > > > + } > > > + > > > return pfn_to_page(pfn); > > > } > > > diff --git a/mm/sparse.c b/mm/sparse.c > > > index 77d91e565045..0c100ae1cf8b 100644 > > > --- a/mm/sparse.c > > > +++ b/mm/sparse.c > > > @@ -14,7 +14,7 @@ > > > #include > > > #include > > > #include > > > - > > > +#include > > > #include "internal.h" > > > #include > > > > > > @@ -465,6 +465,9 @@ static void __init sparse_buffer_init(unsigned > long size, int nid) > > > */ > > > sparsemap_buf =3D memmap_alloc(size, section_map_size(), addr, ni= d, > true); > > > sparsemap_buf_end =3D sparsemap_buf + size; > > > +#ifndef CONFIG_SPARSEMEM_VMEMMAP > > > + mod_node_early_perpage_metadata(nid, DIV_ROUND_UP(size, > PAGE_SIZE)); > > > +#endif > > > } > > > > > > static void __init sparse_buffer_fini(void) > > > @@ -641,6 +644,8 @@ static void depopulate_section_memmap(unsigned > long pfn, unsigned long nr_pages, > > > unsigned long start =3D (unsigned long) pfn_to_page(pfn); > > > unsigned long end =3D start + nr_pages * sizeof(struct page); > > > > > > + mod_node_page_state(page_pgdat(pfn_to_page(pfn)), NR_PAGE_METADAT= A, > > > + -1L * (DIV_ROUND_UP(end - start, PAGE_SIZE)))= ; > > > vmemmap_free(start, end, altmap); > > > } > > > static void free_map_bootmem(struct page *memmap) > > > diff --git a/mm/vmstat.c b/mm/vmstat.c > > > index 359460deb377..23e88d8c21b7 100644 > > > --- a/mm/vmstat.c > > > +++ b/mm/vmstat.c > > > @@ -1249,7 +1249,8 @@ const char * const vmstat_text[] =3D { > > > "pgpromote_success", > > > "pgpromote_candidate", > > > #endif > > > - > > > + "nr_page_metadata", > > > + "nr_page_metadata_boot", > > > /* enum writeback_stat_item counters */ > > > "nr_dirty_threshold", > > > "nr_dirty_background_threshold", > > > @@ -2278,4 +2279,27 @@ static int __init extfrag_debug_init(void) > > > } > > > > > > module_init(extfrag_debug_init); > > > + > > > #endif > > > + > > > +/* > > > + * Page metadata size (struct page and page_ext) in pages > > > + */ > > > +static unsigned long early_perpage_metadata[MAX_NUMNODES] __initdata= ; > > > + > > > +void __init mod_node_early_perpage_metadata(int nid, long delta) > > > +{ > > > + early_perpage_metadata[nid] +=3D delta; > > > +} > > > + > > > +void __init store_early_perpage_metadata(void) > > > +{ > > > + int nid; > > > + struct pglist_data *pgdat; > > > + > > > + for_each_online_pgdat(pgdat) { > > > + nid =3D pgdat->node_id; > > > + mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA_BOOT= , > > > + early_perpage_metadata[nid]); > > > + } > > > +} > > > -- > > > 2.43.0.472.g3155946c3a-goog > > > > > > > > > > > > --000000000000f716e7060e8a651e Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable


=
On Fri, Jan 5, 2024 at 4:38=E2=80=AFP= M David Rientjes <rientjes@google= .com> wrote:
On Fri, 5 Jan 2024, David Rientjes wrote:

> On Tue, 5 Dec 2023, Sourav Panda wrote:
>
> > diff --git a/Documentation/filesystems/proc.rst b/Documentation/f= ilesystems/proc.rst
> > index 49ef12df631b..d5901d04e082 100644
> > --- a/Documentation/filesystems/proc.rst
> > +++ b/Documentation/filesystems/proc.rst
> > @@ -993,6 +993,7 @@ Example output. You may not have all of these= fields.
> >=C2=A0 =C2=A0 =C2=A0 AnonPages:=C2=A0 =C2=A0 =C2=A0 =C2=A04654780 = kB
> >=C2=A0 =C2=A0 =C2=A0 Mapped:=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0266244 kB
> >=C2=A0 =C2=A0 =C2=A0 Shmem:=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 9976 kB
> > +=C2=A0 =C2=A0 PageMetadata:=C2=A0 =C2=A0 =C2=A0513419 kB
> >=C2=A0 =C2=A0 =C2=A0 KReclaimable:=C2=A0 =C2=A0 =C2=A0517708 kB > >=C2=A0 =C2=A0 =C2=A0 Slab:=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0660044 kB
> >=C2=A0 =C2=A0 =C2=A0 SReclaimable:=C2=A0 =C2=A0 =C2=A0517708 kB > > @@ -1095,6 +1096,8 @@ Mapped
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 files whic= h have been mmapped, such as libraries
> >=C2=A0 Shmem
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Total memo= ry used by shared memory (shmem) and tmpfs
> > +PageMetadata
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Memory used for= per-page metadata
> >=C2=A0 KReclaimable
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Kernel all= ocations that the kernel will attempt to reclaim
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 under memo= ry pressure. Includes SReclaimable (below), and other
> > diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
> > index 45af9a989d40..f141bb2a550d 100644
> > --- a/fs/proc/meminfo.c
> > +++ b/fs/proc/meminfo.c
> > @@ -39,7 +39,9 @@ static int meminfo_proc_show(struct seq_file *m= , void *v)
> >=C2=A0 =C2=A0 =C2=A0long available;
> >=C2=A0 =C2=A0 =C2=A0unsigned long pages[NR_LRU_LISTS];
> >=C2=A0 =C2=A0 =C2=A0unsigned long sreclaimable, sunreclaim;
> > +=C2=A0 =C2=A0unsigned long nr_page_metadata;
>
> Initialize it here (if we actually need this variable)?
>

Thanks David! I will make this change = in v7.
=C2=A0

> >=C2=A0 =C2=A0 =C2=A0int lru;
> > +=C2=A0 =C2=A0int nid;
> >=C2=A0
> >=C2=A0 =C2=A0 =C2=A0si_meminfo(&i);
> >=C2=A0 =C2=A0 =C2=A0si_swapinfo(&i);
> > @@ -57,6 +59,10 @@ static int meminfo_proc_show(struct seq_file *= m, void *v)
> >=C2=A0 =C2=A0 =C2=A0sreclaimable =3D global_node_page_state_pages(= NR_SLAB_RECLAIMABLE_B);
> >=C2=A0 =C2=A0 =C2=A0sunreclaim =3D global_node_page_state_pages(NR= _SLAB_UNRECLAIMABLE_B);
> >=C2=A0
> > +=C2=A0 =C2=A0nr_page_metadata =3D 0;
> > +=C2=A0 =C2=A0for_each_online_node(nid)
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0nr_page_metadata +=3D n= ode_page_state(NODE_DATA(nid), NR_PAGE_METADATA);
>
> Is this intended to be different than
> global_node_page_state_pages(NR_PAGE_METADATA)?=C2=A0
>

We also presen= t the metrics in /proc/meminfo to export the value in a readable format whe= re all the values are consistently in kB (unlike vmstat with varying units)= .



> If so, any hint as to why we want to discount page metadata on offline=
> nodes?=C2=A0 We can't make an inference that metadata is always al= located
> locally, memoryless nodes need things like struct page allocated on no= des
> with memory.
>

Sorry, meant a node with only ZONE_MOVABLE here so metadata can't be allocated locally.

= When we have a ZONE_MOVABLE=E2=80=99s page's metadata on another node, = it would NOT be discounted. It would be accounted for on the other node.
=C2=A0

But would be very interested to learn why this subtlety exists to sum up only online nodes.

> So even if a memoryless node is offline, we'd still be including i= ts
> metadata here with the current implementation.
>
> Or maybe I'm missing a subtlety here for why this is not already <= br> > global_node_page_state_pages().
>

If I understoo= d correctly, as long as the memory is not hotplugged, it won't have any= page metadata and thus we would NOT include its metadata. The moment the m= emory is hotplugged or hot-unplugged, the metadata will be incremented or d= ecremented, correspondingly.
=C2=A0

> > +
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "MemTotal:=C2=A0 =C2=A0 = =C2=A0 =C2=A0", i.totalram);
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "MemFree:=C2=A0 =C2=A0 =C2= =A0 =C2=A0 ", i.freeram);
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "MemAvailable:=C2=A0 =C2= =A0", available);
> > @@ -104,6 +110,7 @@ static int meminfo_proc_show(struct seq_file = *m, void *v)
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "Mapped:=C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0",
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0glob= al_node_page_state(NR_FILE_MAPPED));
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "Shmem:=C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 ", i.sharedram);
> > +=C2=A0 =C2=A0show_val_kb(m, "PageMetadata:=C2=A0 =C2=A0&quo= t;, nr_page_metadata);
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "KReclaimable:=C2=A0 =C2= =A0", sreclaimable +
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0glob= al_node_page_state(NR_KERNEL_MISC_RECLAIMABLE));
> >=C2=A0 =C2=A0 =C2=A0show_val_kb(m, "Slab:=C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0", sreclaimable + sunreclaim);
> > diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> > index 3c25226beeed..ef176152be7c 100644
> > --- a/include/linux/mmzone.h
> > +++ b/include/linux/mmzone.h
> > @@ -207,6 +207,10 @@ enum node_stat_item {
> >=C2=A0 =C2=A0 =C2=A0PGPROMOTE_SUCCESS,=C2=A0 =C2=A0 =C2=A0 /* prom= ote successfully */
> >=C2=A0 =C2=A0 =C2=A0PGPROMOTE_CANDIDATE,=C2=A0 =C2=A0 /* candidate= pages to promote */
> >=C2=A0 #endif
> > +=C2=A0 =C2=A0NR_PAGE_METADATA,=C2=A0 =C2=A0 =C2=A0 =C2=A0/* Page= metadata size (struct page and page_ext)
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 * in pages
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 */
> > +=C2=A0 =C2=A0NR_PAGE_METADATA_BOOT,=C2=A0 /* NR_PAGE_METADATA fo= r bootmem */
>
> So if some vmemmap pages are freed, then MemTotal could be incremented= by
> a portion of NR_PAGE_METADATA_BOOT and then this stat is decremented?= =C2=A0 Is
> the goal that the sum of MemTotal + SUM(nr_page_metadata_boot) is alwa= ys
> constant?
>

Yep, this is correct.
=C2=A0=

> >=C2=A0 =C2=A0 =C2=A0NR_VM_NODE_STAT_ITEMS
> >=C2=A0 };
> >=C2=A0
> > diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
> > index fed855bae6d8..af096a881f03 100644
> > --- a/include/linux/vmstat.h
> > +++ b/include/linux/vmstat.h
> > @@ -656,4 +656,8 @@ static inline void lruvec_stat_sub_folio(stru= ct folio *folio,
> >=C2=A0 {
> >=C2=A0 =C2=A0 =C2=A0lruvec_stat_mod_folio(folio, idx, -folio_nr_pa= ges(folio));
> >=C2=A0 }
> > +
> > +void __init mod_node_early_perpage_metadata(int nid, long delta)= ;
> > +void __init store_early_perpage_metadata(void);
> > +
> >=C2=A0 #endif /* _LINUX_VMSTAT_H */
> > diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c
> > index 87818ee7f01d..5b10d8d2b471 100644
> > --- a/mm/hugetlb_vmemmap.c
> > +++ b/mm/hugetlb_vmemmap.c
> > @@ -230,10 +230,14 @@ static int vmemmap_remap_range(unsigned lon= g start, unsigned long end,
> >=C2=A0 =C2=A0*/
> >=C2=A0 static inline void free_vmemmap_page(struct page *page)
> >=C2=A0 {
> > -=C2=A0 =C2=A0if (PageReserved(page))
> > +=C2=A0 =C2=A0if (PageReserved(page)) {
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0free_bootmem_page(= page);
> > -=C2=A0 =C2=A0else
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(pag= e_pgdat(page), NR_PAGE_METADATA_BOOT,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0-1);
> > +=C2=A0 =C2=A0} else {
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0__free_page(page);=
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(pag= e_pgdat(page), NR_PAGE_METADATA, -1);
> > +=C2=A0 =C2=A0}
> >=C2=A0 }
> >=C2=A0
> >=C2=A0 /* Free a list of the vmemmap pages */
> > @@ -389,6 +393,7 @@ static int vmemmap_remap_free(unsigned long s= tart, unsigned long end,
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0copy_page(page_to_= virt(walk.reuse_page),
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0(void *)walk.reuse_addr);
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0list_add(&walk= .reuse_page->lru, vmemmap_pages);
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(NOD= E_DATA(nid), NR_PAGE_METADATA, 1);
> >=C2=A0 =C2=A0 =C2=A0}
> >=C2=A0
> >=C2=A0 =C2=A0 =C2=A0/*
> > @@ -437,14 +442,20 @@ static int alloc_vmemmap_page_list(unsigned= long start, unsigned long end,
> >=C2=A0 =C2=A0 =C2=A0unsigned long nr_pages =3D (end - start) >&= gt; PAGE_SHIFT;
> >=C2=A0 =C2=A0 =C2=A0int nid =3D page_to_nid((struct page *)start);=
> >=C2=A0 =C2=A0 =C2=A0struct page *page, *next;
> > +=C2=A0 =C2=A0int i;
> >=C2=A0
> > -=C2=A0 =C2=A0while (nr_pages--) {
> > +=C2=A0 =C2=A0for (i =3D 0; i < nr_pages; i++) {
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0page =3D alloc_pag= es_node(nid, gfp_mask, 0);
> > -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0if (!page)
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0if (!page) {
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADATA,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0i);
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0goto out;
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0}
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0list_add(&page= ->lru, list);
> >=C2=A0 =C2=A0 =C2=A0}
> >=C2=A0
> > +=C2=A0 =C2=A0mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADAT= A, nr_pages);
> > +
> >=C2=A0 =C2=A0 =C2=A0return 0;
> >=C2=A0 out:
> >=C2=A0 =C2=A0 =C2=A0list_for_each_entry_safe(page, next, list, lru= )
> > diff --git a/mm/mm_init.c b/mm/mm_init.c
> > index 077bfe393b5e..38f8e1f454a0 100644
> > --- a/mm/mm_init.c
> > +++ b/mm/mm_init.c
> > @@ -26,6 +26,7 @@
> >=C2=A0 #include <linux/pgtable.h>
> >=C2=A0 #include <linux/swap.h>
> >=C2=A0 #include <linux/cma.h>
> > +#include <linux/vmstat.h>
> >=C2=A0 #include "internal.h"
> >=C2=A0 #include "slab.h"
> >=C2=A0 #include "shuffle.h"
> > @@ -1656,6 +1657,8 @@ static void __init alloc_node_mem_map(struc= t pglist_data *pgdat)
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0panic("Failed to allocate %ld bytes for node %d memory map\n= ",
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0size, pgdat->node_id);
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0pgdat->node_mem= _map =3D map + offset;
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_early_perpage_= metadata(pgdat->node_id,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0DIV_ROUND_UP(size, PAGE_SIZE));
> >=C2=A0 =C2=A0 =C2=A0}
> >=C2=A0 =C2=A0 =C2=A0pr_debug("%s: node %d, pgdat %08lx, node_= mem_map %08lx\n",
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0__func__, pgdat->node_id, (unsigne= d long)pgdat,
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index 733732e7e0ba..dd78017105b0 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -5636,6 +5636,7 @@ void __init setup_per_cpu_pageset(void)
> >=C2=A0 =C2=A0 =C2=A0for_each_online_pgdat(pgdat)
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0pgdat->per_cpu_= nodestats =3D
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0alloc_percpu(struct per_cpu_nodestat);
> > +=C2=A0 =C2=A0store_early_perpage_metadata();
> >=C2=A0 }
> >=C2=A0
> >=C2=A0 __meminit void zone_pcp_init(struct zone *zone)
> > diff --git a/mm/page_ext.c b/mm/page_ext.c
> > index 4548fcc66d74..4ca9f298f34e 100644
> > --- a/mm/page_ext.c
> > +++ b/mm/page_ext.c
> > @@ -201,6 +201,8 @@ static int __init alloc_node_page_ext(int nid= )
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0return -ENOMEM; > >=C2=A0 =C2=A0 =C2=A0NODE_DATA(nid)->node_page_ext =3D base;
> >=C2=A0 =C2=A0 =C2=A0total_usage +=3D table_size;
> > +=C2=A0 =C2=A0mod_node_page_state(NODE_DATA(nid), NR_PAGE_METADAT= A_BOOT,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0DIV_ROUND_UP(table_size, PAGE_SIZE));
> >=C2=A0 =C2=A0 =C2=A0return 0;
> >=C2=A0 }
> >=C2=A0
> > @@ -255,12 +257,15 @@ static void *__meminit alloc_page_ext(size_= t size, int nid)
> >=C2=A0 =C2=A0 =C2=A0void *addr =3D NULL;
> >=C2=A0
> >=C2=A0 =C2=A0 =C2=A0addr =3D alloc_pages_exact_nid(nid, size, flag= s);
> > -=C2=A0 =C2=A0if (addr) {
> > +=C2=A0 =C2=A0if (addr)
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0kmemleak_alloc(add= r, size, 1, flags);
> > -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0return addr;
> > -=C2=A0 =C2=A0}
> > +=C2=A0 =C2=A0else
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0addr =3D vzalloc_node(s= ize, nid);
> >=C2=A0
> > -=C2=A0 =C2=A0addr =3D vzalloc_node(size, nid);
> > +=C2=A0 =C2=A0if (addr) {
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(NOD= E_DATA(nid), NR_PAGE_METADATA,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0DIV_ROUND_UP(size, PAGE_SIZ= E));
> > +=C2=A0 =C2=A0}
> >=C2=A0
> >=C2=A0 =C2=A0 =C2=A0return addr;
> >=C2=A0 }
> > @@ -303,18 +308,27 @@ static int __meminit init_section_page_ext(= unsigned long pfn, int nid)
> >=C2=A0
> >=C2=A0 static void free_page_ext(void *addr)
> >=C2=A0 {
> > +=C2=A0 =C2=A0size_t table_size;
> > +=C2=A0 =C2=A0struct page *page;
> > +=C2=A0 =C2=A0struct pglist_data *pgdat;
> > +
> > +=C2=A0 =C2=A0table_size =3D page_ext_size * PAGES_PER_SECTION; > > +
> >=C2=A0 =C2=A0 =C2=A0if (is_vmalloc_addr(addr)) {
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0page =3D vmalloc_to_pag= e(addr);
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0pgdat =3D page_pgdat(pa= ge);
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0vfree(addr);
> >=C2=A0 =C2=A0 =C2=A0} else {
> > -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0struct page *page =3D v= irt_to_page(addr);
> > -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0size_t table_size;
> > -
> > -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0table_size =3D page_ext= _size * PAGES_PER_SECTION;
> > -
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0page =3D virt_to_page(a= ddr);
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0pgdat =3D page_pgdat(pa= ge);
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0BUG_ON(PageReserve= d(page));
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0kmemleak_free(addr= );
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0free_pages_exact(a= ddr, table_size);
> >=C2=A0 =C2=A0 =C2=A0}
> > +
> > +=C2=A0 =C2=A0mod_node_page_state(pgdat, NR_PAGE_METADATA,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0-1L * (DIV_ROUND_UP(table_size, PAGE_SIZE)));
> > +
> >=C2=A0 }
> >=C2=A0
> >=C2=A0 static void __free_page_ext(unsigned long pfn)
> > diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c
> > index a2cbe44c48e1..054b49539843 100644
> > --- a/mm/sparse-vmemmap.c
> > +++ b/mm/sparse-vmemmap.c
> > @@ -469,5 +469,13 @@ struct page * __meminit __populate_section_m= emmap(unsigned long pfn,
> >=C2=A0 =C2=A0 =C2=A0if (r < 0)
> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0return NULL;
> >=C2=A0
> > +=C2=A0 =C2=A0if (system_state =3D=3D SYSTEM_BOOTING) {
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(NOD= E_DATA(nid), NR_PAGE_METADATA_BOOT,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0DIV_ROUND_UP(end - start, P= AGE_SIZE));
> > +=C2=A0 =C2=A0} else {
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(NOD= E_DATA(nid), NR_PAGE_METADATA,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0DIV_ROUND_UP(end - start, P= AGE_SIZE));
> > +=C2=A0 =C2=A0}
> > +
> >=C2=A0 =C2=A0 =C2=A0return pfn_to_page(pfn);
> >=C2=A0 }
> > diff --git a/mm/sparse.c b/mm/sparse.c
> > index 77d91e565045..0c100ae1cf8b 100644
> > --- a/mm/sparse.c
> > +++ b/mm/sparse.c
> > @@ -14,7 +14,7 @@
> >=C2=A0 #include <linux/swap.h>
> >=C2=A0 #include <linux/swapops.h>
> >=C2=A0 #include <linux/bootmem_info.h>
> > -
> > +#include <linux/vmstat.h>
> >=C2=A0 #include "internal.h"
> >=C2=A0 #include <asm/dma.h>
> >=C2=A0
> > @@ -465,6 +465,9 @@ static void __init sparse_buffer_init(unsigne= d long size, int nid)
> >=C2=A0 =C2=A0 =C2=A0 */
> >=C2=A0 =C2=A0 =C2=A0sparsemap_buf =3D memmap_alloc(size, section_m= ap_size(), addr, nid, true);
> >=C2=A0 =C2=A0 =C2=A0sparsemap_buf_end =3D sparsemap_buf + size; > > +#ifndef CONFIG_SPARSEMEM_VMEMMAP
> > +=C2=A0 =C2=A0mod_node_early_perpage_metadata(nid, DIV_ROUND_UP(s= ize, PAGE_SIZE));
> > +#endif
> >=C2=A0 }
> >=C2=A0
> >=C2=A0 static void __init sparse_buffer_fini(void)
> > @@ -641,6 +644,8 @@ static void depopulate_section_memmap(unsigne= d long pfn, unsigned long nr_pages,
> >=C2=A0 =C2=A0 =C2=A0unsigned long start =3D (unsigned long) pfn_to= _page(pfn);
> >=C2=A0 =C2=A0 =C2=A0unsigned long end =3D start + nr_pages * sizeo= f(struct page);
> >=C2=A0
> > +=C2=A0 =C2=A0mod_node_page_state(page_pgdat(pfn_to_page(pfn)), N= R_PAGE_METADATA,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0-1L * (DIV_ROUND_UP(end - start, PAGE_SIZE)));
> >=C2=A0 =C2=A0 =C2=A0vmemmap_free(start, end, altmap);
> >=C2=A0 }
> >=C2=A0 static void free_map_bootmem(struct page *memmap)
> > diff --git a/mm/vmstat.c b/mm/vmstat.c
> > index 359460deb377..23e88d8c21b7 100644
> > --- a/mm/vmstat.c
> > +++ b/mm/vmstat.c
> > @@ -1249,7 +1249,8 @@ const char * const vmstat_text[] =3D {
> >=C2=A0 =C2=A0 =C2=A0"pgpromote_success",
> >=C2=A0 =C2=A0 =C2=A0"pgpromote_candidate",
> >=C2=A0 #endif
> > -
> > +=C2=A0 =C2=A0"nr_page_metadata",
> > +=C2=A0 =C2=A0"nr_page_metadata_boot",
> >=C2=A0 =C2=A0 =C2=A0/* enum writeback_stat_item counters */
> >=C2=A0 =C2=A0 =C2=A0"nr_dirty_threshold",
> >=C2=A0 =C2=A0 =C2=A0"nr_dirty_background_threshold",
> > @@ -2278,4 +2279,27 @@ static int __init extfrag_debug_init(void)=
> >=C2=A0 }
> >=C2=A0
> >=C2=A0 module_init(extfrag_debug_init);
> > +
> >=C2=A0 #endif
> > +
> > +/*
> > + * Page metadata size (struct page and page_ext) in pages
> > + */
> > +static unsigned long early_perpage_metadata[MAX_NUMNODES] __init= data;
> > +
> > +void __init mod_node_early_perpage_metadata(int nid, long delta)=
> > +{
> > +=C2=A0 =C2=A0early_perpage_metadata[nid] +=3D delta;
> > +}
> > +
> > +void __init store_early_perpage_metadata(void)
> > +{
> > +=C2=A0 =C2=A0int nid;
> > +=C2=A0 =C2=A0struct pglist_data *pgdat;
> > +
> > +=C2=A0 =C2=A0for_each_online_pgdat(pgdat) {
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0nid =3D pgdat->node_= id;
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0mod_node_page_state(NOD= E_DATA(nid), NR_PAGE_METADATA_BOOT,
> > +=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0early_perpage_metadata[nid]= );
> > +=C2=A0 =C2=A0}
> > +}
> > --
> > 2.43.0.472.g3155946c3a-goog
> >
> >
> >
>
--000000000000f716e7060e8a651e--