mm: memcontrol: convert NR_ANON_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 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_ANON_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/20201228164110.2838-3-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Rafael. J. Wysocki <rafael@kernel.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Roman Gushchin <guro@fb.com>
Cc: Sami Tolvanen <samitolvanen@google.com>
Cc: Feng Tang <feng.tang@intel.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Pankaj Gupta <pankaj.gupta@cloud.ionos.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

8 files changed, 44 insertions, 28 deletions

diff --git a/drivers/base/node.c b/drivers/base/node.c
index 04f71c7bc3f8..6da0c3508bc9 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -461,8 +461,7 @@ static ssize_t node_read_meminfo(struct device *dev,
 			     nid, K(sunreclaimable)
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 			     ,
-			     nid, K(node_page_state(pgdat, NR_ANON_THPS) *
-				    HPAGE_PMD_NR),
+			     nid, K(node_page_state(pgdat, NR_ANON_THPS)),
 			     nid, K(node_page_state(pgdat, NR_SHMEM_THPS) *
 				    HPAGE_PMD_NR),
 			     nid, K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED) *
@@ -519,10 +518,14 @@ static ssize_t node_read_vmstat(struct device *dev,
 				     sum_zone_numa_state(nid, i));
 
 #endif
-	for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++)
-		len += sysfs_emit_at(buf, len, "%s %lu\n",
-				     node_stat_name(i),
-				     node_page_state_pages(pgdat, i));
+	for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) {
+		unsigned long pages = node_page_state_pages(pgdat, i);
+
+		if (vmstat_item_print_in_thp(i))
+			pages /= HPAGE_PMD_NR;
+		len += sysfs_emit_at(buf, len, "%s %lu\n", node_stat_name(i),
+				     pages);
+	}
 
 	return len;
 }
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index d6fc74619625..a635c8a84ddf 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -129,7 +129,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 	show_val_kb(m, "AnonHugePages:  ",
-		    global_node_page_state(NR_ANON_THPS) * HPAGE_PMD_NR);
+		    global_node_page_state(NR_ANON_THPS));
 	show_val_kb(m, "ShmemHugePages: ",
 		    global_node_page_state(NR_SHMEM_THPS) * HPAGE_PMD_NR);
 	show_val_kb(m, "ShmemPmdMapped: ",
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index b593316bff3d..67d50ef5dd20 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -210,6 +210,19 @@ enum node_stat_item {
 };
 
 /*
+ * Returns true if the item should be printed in THPs (/proc/vmstat
+ * currently prints number of anon, file and shmem THPs. But the item
+ * is charged in pages).
+ */
+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;
+}
+
+/*
  * Returns true if the value is measured in bytes (most vmstat values are
  * measured in pages). This defines the API part, the internal representation
  * might be different.
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index f655137536eb..3691e863070a 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2176,7 +2176,8 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 		lock_page_memcg(page);
 		if (atomic_add_negative(-1, compound_mapcount_ptr(page))) {
 			/* Last compound_mapcount is gone. */
-			__dec_lruvec_page_state(page, NR_ANON_THPS);
+			__mod_lruvec_page_state(page, NR_ANON_THPS,
+						-HPAGE_PMD_NR);
 			if (TestClearPageDoubleMap(page)) {
 				/* No need in mapcount reference anymore */
 				for (i = 0; i < HPAGE_PMD_NR; i++)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e31e47e7bab2..b2405f049006 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1533,7 +1533,7 @@ static struct memory_stat memory_stats[] = {
 	 * on some architectures, the macro of HPAGE_PMD_SIZE is not
 	 * constant(e.g. powerpc).
 	 */
-	{ "anon_thp", 0, NR_ANON_THPS },
+	{ "anon_thp", PAGE_SIZE, NR_ANON_THPS },
 	{ "file_thp", 0, NR_FILE_THPS },
 	{ "shmem_thp", 0, NR_SHMEM_THPS },
 #endif
@@ -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_ANON_THPS ||
-		    memory_stats[i].idx == NR_FILE_THPS ||
+		if (memory_stats[i].idx == NR_FILE_THPS ||
 		    memory_stats[i].idx == NR_SHMEM_THPS)
 			memory_stats[i].ratio = HPAGE_PMD_SIZE;
 #endif
@@ -4087,10 +4086,6 @@ static int memcg_stat_show(struct seq_file *m, void *v)
 		if (memcg1_stats[i] == MEMCG_SWAP && !do_memsw_account())
 			continue;
 		nr = memcg_page_state_local(memcg, memcg1_stats[i]);
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-		if (memcg1_stats[i] == NR_ANON_THPS)
-			nr *= HPAGE_PMD_NR;
-#endif
 		seq_printf(m, "%s %lu\n", memcg1_stat_names[i], nr * PAGE_SIZE);
 	}
 
@@ -4121,10 +4116,6 @@ static int memcg_stat_show(struct seq_file *m, void *v)
 		if (memcg1_stats[i] == MEMCG_SWAP && !do_memsw_account())
 			continue;
 		nr = memcg_page_state(memcg, memcg1_stats[i]);
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-		if (memcg1_stats[i] == NR_ANON_THPS)
-			nr *= HPAGE_PMD_NR;
-#endif
 		seq_printf(m, "total_%s %llu\n", memcg1_stat_names[i],
 						(u64)nr * PAGE_SIZE);
 	}
@@ -5652,10 +5643,11 @@ static int mem_cgroup_move_account(struct page *page,
 			__mod_lruvec_state(from_vec, NR_ANON_MAPPED, -nr_pages);
 			__mod_lruvec_state(to_vec, NR_ANON_MAPPED, nr_pages);
 			if (PageTransHuge(page)) {
-				__dec_lruvec_state(from_vec, NR_ANON_THPS);
-				__inc_lruvec_state(to_vec, NR_ANON_THPS);
+				__mod_lruvec_state(from_vec, NR_ANON_THPS,
+						   -nr_pages);
+				__mod_lruvec_state(to_vec, NR_ANON_THPS,
+						   nr_pages);
 			}
-
 		}
 	} else {
 		__mod_lruvec_state(from_vec, NR_FILE_PAGES, -nr_pages);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ef5070fed76b..2e2e47f8714b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5587,7 +5587,7 @@ void show_free_areas(unsigned int filter, nodemask_t *nodemask)
 			K(node_page_state(pgdat, NR_SHMEM_THPS) * HPAGE_PMD_NR),
 			K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED)
 					* HPAGE_PMD_NR),
-			K(node_page_state(pgdat, NR_ANON_THPS) * HPAGE_PMD_NR),
+			K(node_page_state(pgdat, NR_ANON_THPS)),
 #endif
 			K(node_page_state(pgdat, NR_WRITEBACK_TEMP)),
 			node_page_state(pgdat, NR_KERNEL_STACK_KB),
diff --git a/mm/rmap.c b/mm/rmap.c
index 08c56aaf72eb..c4d5c63cfd29 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1144,7 +1144,7 @@ void do_page_add_anon_rmap(struct page *page,
 		 * disabled.
 		 */
 		if (compound)
-			__inc_lruvec_page_state(page, NR_ANON_THPS);
+			__mod_lruvec_page_state(page, NR_ANON_THPS, nr);
 		__mod_lruvec_page_state(page, NR_ANON_MAPPED, nr);
 	}
 
@@ -1186,7 +1186,7 @@ void page_add_new_anon_rmap(struct page *page,
 		if (hpage_pincount_available(page))
 			atomic_set(compound_pincount_ptr(page), 0);
 
-		__inc_lruvec_page_state(page, NR_ANON_THPS);
+		__mod_lruvec_page_state(page, NR_ANON_THPS, nr);
 	} else {
 		/* Anon THP always mapped first with PMD */
 		VM_BUG_ON_PAGE(PageTransCompound(page), page);
@@ -1292,7 +1292,7 @@ static void page_remove_anon_compound_rmap(struct page *page)
 	if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE))
 		return;
 
-	__dec_lruvec_page_state(page, NR_ANON_THPS);
+	__mod_lruvec_page_state(page, NR_ANON_THPS, -thp_nr_pages(page));
 
 	if (TestClearPageDoubleMap(page)) {
 		/*
diff --git a/mm/vmstat.c b/mm/vmstat.c
index f8942160fc95..07dc0af50cf0 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1619,8 +1619,12 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat,
 	if (is_zone_first_populated(pgdat, zone)) {
 		seq_printf(m, "\n  per-node stats");
 		for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) {
+			unsigned long pages = node_page_state_pages(pgdat, i);
+
+			if (vmstat_item_print_in_thp(i))
+				pages /= HPAGE_PMD_NR;
 			seq_printf(m, "\n      %-12s %lu", node_stat_name(i),
-				   node_page_state_pages(pgdat, i));
+				   pages);
 		}
 	}
 	seq_printf(m,
@@ -1740,8 +1744,11 @@ static void *vmstat_start(struct seq_file *m, loff_t *pos)
 	v += NR_VM_NUMA_STAT_ITEMS;
 #endif
 
-	for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++)
+	for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) {
 		v[i] = global_node_page_state_pages(i);
+		if (vmstat_item_print_in_thp(i))
+			v[i] /= HPAGE_PMD_NR;
+	}
 	v += NR_VM_NODE_STAT_ITEMS;
 
 	global_dirty_limits(v + NR_DIRTY_BG_THRESHOLD,