mm: memcontrol: convert NR_SHMEM_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_SHMEM_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/20201228164110.2838-5-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>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

6 files changed, 8 insertions, 29 deletions

diff --git a/mm/filemap.c b/mm/filemap.c
index 9efe059e2b58..46a8b9e82434 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -206,7 +206,7 @@ static void unaccount_page_cache_page(struct address_space *mapping,
 	if (PageSwapBacked(page)) {
 		__mod_lruvec_page_state(page, NR_SHMEM, -nr);
 		if (PageTransHuge(page))
-			__dec_lruvec_page_state(page, NR_SHMEM_THPS);
+			__mod_lruvec_page_state(page, NR_SHMEM_THPS, -nr);
 	} else if (PageTransHuge(page)) {
 		__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 77181faa2340..86a3015ba54f 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2755,7 +2755,8 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 			int nr = thp_nr_pages(head);
 
 			if (PageSwapBacked(head))
-				__dec_lruvec_page_state(head, NR_SHMEM_THPS);
+				__mod_lruvec_page_state(head, NR_SHMEM_THPS,
+							-nr);
 			else
 				__mod_lruvec_page_state(head, NR_FILE_THPS,
 							-nr);
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index c427fe2ca7ff..75e246f680f4 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -1858,7 +1858,7 @@ out_unlock:
 	nr = thp_nr_pages(new_page);
 
 	if (is_shmem)
-		__inc_lruvec_page_state(new_page, NR_SHMEM_THPS);
+		__mod_lruvec_page_state(new_page, NR_SHMEM_THPS, nr);
 	else {
 		__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 fc552f57d3fb..d3a0c59210e7 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1516,7 +1516,7 @@ struct memory_stat {
 	unsigned int idx;
 };
 
-static struct memory_stat memory_stats[] = {
+static const struct memory_stat memory_stats[] = {
 	{ "anon", PAGE_SIZE, NR_ANON_MAPPED },
 	{ "file", PAGE_SIZE, NR_FILE_PAGES },
 	{ "kernel_stack", 1024, NR_KERNEL_STACK_KB },
@@ -1528,14 +1528,9 @@ static struct memory_stat memory_stats[] = {
 	{ "file_dirty", PAGE_SIZE, NR_FILE_DIRTY },
 	{ "file_writeback", PAGE_SIZE, NR_WRITEBACK },
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	/*
-	 * The ratio will be initialized in memory_stats_init(). Because
-	 * on some architectures, the macro of HPAGE_PMD_SIZE is not
-	 * constant(e.g. powerpc).
-	 */
 	{ "anon_thp", PAGE_SIZE, NR_ANON_THPS },
 	{ "file_thp", PAGE_SIZE, NR_FILE_THPS },
-	{ "shmem_thp", 0, NR_SHMEM_THPS },
+	{ "shmem_thp", PAGE_SIZE, NR_SHMEM_THPS },
 #endif
 	{ "inactive_anon", PAGE_SIZE, NR_INACTIVE_ANON },
 	{ "active_anon", PAGE_SIZE, NR_ACTIVE_ANON },
@@ -1560,23 +1555,6 @@ static struct memory_stat memory_stats[] = {
 	{ "workingset_nodereclaim", 1, WORKINGSET_NODERECLAIM },
 };
 
-static int __init memory_stats_init(void)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(memory_stats); i++) {
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-		if (memory_stats[i].idx == NR_SHMEM_THPS)
-			memory_stats[i].ratio = HPAGE_PMD_SIZE;
-#endif
-		VM_BUG_ON(!memory_stats[i].ratio);
-		VM_BUG_ON(memory_stats[i].idx >= MEMCG_NR_STAT);
-	}
-
-	return 0;
-}
-pure_initcall(memory_stats_init);
-
 static char *memory_stat_format(struct mem_cgroup *memcg)
 {
 	struct seq_buf s;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 2e2e47f8714b..df292d8e659b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5584,7 +5584,7 @@ void show_free_areas(unsigned int filter, nodemask_t *nodemask)
 			K(node_page_state(pgdat, NR_WRITEBACK)),
 			K(node_page_state(pgdat, NR_SHMEM)),
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-			K(node_page_state(pgdat, NR_SHMEM_THPS) * HPAGE_PMD_NR),
+			K(node_page_state(pgdat, NR_SHMEM_THPS)),
 			K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED)
 					* HPAGE_PMD_NR),
 			K(node_page_state(pgdat, NR_ANON_THPS)),
diff --git a/mm/shmem.c b/mm/shmem.c
index 7924b3bf46fb..ff741d229701 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -713,7 +713,7 @@ next:
 		}
 		if (PageTransHuge(page)) {
 			count_vm_event(THP_FILE_ALLOC);
-			__inc_lruvec_page_state(page, NR_SHMEM_THPS);
+			__mod_lruvec_page_state(page, NR_SHMEM_THPS, nr);
 		}
 		mapping->nrpages += nr;
 		__mod_lruvec_page_state(page, NR_FILE_PAGES, nr);