mm: memcontrol: convert NR_FILE_THPS account to pages

Currently we use struct per_cpu_nodestat to cache the vmstat counters,
which leads to inaccurate statistics especially THP vmstat counters.  In
the systems with if hundreds of processors it can be GBs of memory.  For
example, for a 96 CPUs system, the threshold is the maximum number of 125.
And the per cpu counters can cache 23.4375 GB in total.

The THP page is already a form of batched addition (it will add 512 worth
of memory in one go) so skipping the batching seems like sensible.
Although every THP stats update overflows the per-cpu counter, resorting
to atomic global updates.  But it can make the statistics more accuracy
for the THP vmstat counters.

So we convert the NR_FILE_THPS account to pages.  This patch is consistent
with 8f182270dfec ("mm/swap.c: flush lru pvecs on compound page arrival").
Doing this also can make the unit of vmstat counters more unified.
Finally, the unit of the vmstat counters are pages, kB and bytes.  The
B/KB suffix can tell us that the unit is bytes or kB.  The rest which is
without suffix are pages.

Link: https://lkml.kernel.org/r/20201228164110.2838-4-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Feng Tang <feng.tang@intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Pankaj Gupta <pankaj.gupta@cloud.ionos.com>
Cc: Rafael. J. Wysocki <rafael@kernel.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Sami Tolvanen <samitolvanen@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/drivers/base/node.c b/drivers/base/node.c
index 6da0c3508bc9cef6a5060cf297096be66926dbb8..d5952f7549115534359fad78e38a04347a106284 100644 (file)
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -466,8 +466,7 @@ static ssize_t node_read_meminfo(struct device *dev,
                                    HPAGE_PMD_NR),
                             nid, K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED) *
                                    HPAGE_PMD_NR),
-                            nid, K(node_page_state(pgdat, NR_FILE_THPS) *
-                                   HPAGE_PMD_NR),
+                            nid, K(node_page_state(pgdat, NR_FILE_THPS)),
                             nid, K(node_page_state(pgdat, NR_FILE_PMDMAPPED) *
                                    HPAGE_PMD_NR)
 #endif

diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index a635c8a84ddf50254eb046e03d6953838fddee06..7ea4679880c8dfdf197fe621ce056befedc8a2a1 100644 (file)
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -135,7 +135,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
        show_val_kb(m, "ShmemPmdMapped: ",
                    global_node_page_state(NR_SHMEM_PMDMAPPED) * HPAGE_PMD_NR);
        show_val_kb(m, "FileHugePages:  ",
-                   global_node_page_state(NR_FILE_THPS) * HPAGE_PMD_NR);
+                   global_node_page_state(NR_FILE_THPS));
        show_val_kb(m, "FilePmdMapped:  ",
                    global_node_page_state(NR_FILE_PMDMAPPED) * HPAGE_PMD_NR);
 #endif

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 67d50ef5dd2050baf9da4f4de92ca7bf0b4358fe..b751a9898bb6895989b523d152b0350f3f0728e5 100644 (file)
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -219,7 +219,8 @@ static __always_inline bool vmstat_item_print_in_thp(enum node_stat_item item)
        if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE))
                return false;
 
-       return item == NR_ANON_THPS;
+       return item == NR_ANON_THPS ||
+              item == NR_FILE_THPS;
 }
 
 /*

diff --git a/mm/filemap.c b/mm/filemap.c
index 5c9d564317a5c5eb2e7f5c4edc6e5e3571bd9c2d..c7527f7607123663889348124d2171357e187b03 100644 (file)
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -206,7 +206,7 @@ static void unaccount_page_cache_page(struct address_space *mapping,
                if (PageTransHuge(page))
                        __dec_lruvec_page_state(page, NR_SHMEM_THPS);
        } else if (PageTransHuge(page)) {
-               __dec_lruvec_page_state(page, NR_FILE_THPS);
+               __mod_lruvec_page_state(page, NR_FILE_THPS, -nr);
                filemap_nr_thps_dec(mapping);
        }
 

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 512ec55376258b4ffd8a18fd98e6202c33943e72..5f2c9f676bd777b37642374e7870e4c588efe0f0 100644 (file)
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2743,10 +2743,13 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
                }
                spin_unlock(&ds_queue->split_queue_lock);
                if (mapping) {
+                       int nr = thp_nr_pages(head);
+
                        if (PageSwapBacked(head))
                                __dec_lruvec_page_state(head, NR_SHMEM_THPS);
                        else
-                               __dec_lruvec_page_state(head, NR_FILE_THPS);
+                               __mod_lruvec_page_state(head, NR_FILE_THPS,
+                                                       -nr);
                }
 
                __split_huge_page(page, list, end);

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 67ab391a537393251b22591a310118f76525438d..ca57dec62f26f0a37e79e5130ee1113c6078e145 100644 (file)
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -1644,6 +1644,7 @@ static void collapse_file(struct mm_struct *mm,
        XA_STATE_ORDER(xas, &mapping->i_pages, start, HPAGE_PMD_ORDER);
        int nr_none = 0, result = SCAN_SUCCEED;
        bool is_shmem = shmem_file(file);
+       int nr;
 
        VM_BUG_ON(!IS_ENABLED(CONFIG_READ_ONLY_THP_FOR_FS) && !is_shmem);
        VM_BUG_ON(start & (HPAGE_PMD_NR - 1));
@@ -1855,11 +1856,12 @@ out_unlock:
                put_page(page);
                goto xa_unlocked;
        }
+       nr = thp_nr_pages(new_page);
 
        if (is_shmem)
                __inc_lruvec_page_state(new_page, NR_SHMEM_THPS);
        else {
-               __inc_lruvec_page_state(new_page, NR_FILE_THPS);
+               __mod_lruvec_page_state(new_page, NR_FILE_THPS, nr);
                filemap_nr_thps_inc(mapping);
        }
 

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 9bce6a9c3759e12912f8a035edfcfb4b2cd07d6d..cb3e495920a9911f325756e864c1574b5fa3026e 100644 (file)
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1534,7 +1534,7 @@ static struct memory_stat memory_stats[] = {
         * constant(e.g. powerpc).
         */
        { "anon_thp", PAGE_SIZE, NR_ANON_THPS },
-       { "file_thp", 0, NR_FILE_THPS },
+       { "file_thp", PAGE_SIZE, NR_FILE_THPS },
        { "shmem_thp", 0, NR_SHMEM_THPS },
 #endif
        { "inactive_anon", PAGE_SIZE, NR_INACTIVE_ANON },
@@ -1566,8 +1566,7 @@ static int __init memory_stats_init(void)
 
        for (i = 0; i < ARRAY_SIZE(memory_stats); i++) {
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-               if (memory_stats[i].idx == NR_FILE_THPS ||
-                   memory_stats[i].idx == NR_SHMEM_THPS)
+               if (memory_stats[i].idx == NR_SHMEM_THPS)
                        memory_stats[i].ratio = HPAGE_PMD_SIZE;
 #endif
                VM_BUG_ON(!memory_stats[i].ratio);