+ mm-base-lru-balancing-on-an-explicit-cost-model.patch added to -mm tree

[Andrew Morton <akpm@linux-foundation.org>]

The patch titled
     Subject: mm: base LRU balancing on an explicit cost model
has been added to the -mm tree.  Its filename is
     mm-base-lru-balancing-on-an-explicit-cost-model.patch

This patch should soon appear at
    http://ozlabs.org/~akpm/mmots/broken-out/mm-base-lru-balancing-on-an-explicit-cost-model.patch
and later at
    http://ozlabs.org/~akpm/mmotm/broken-out/mm-base-lru-balancing-on-an-explicit-cost-model.patch

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***

The -mm tree is included into linux-next and is updated
there every 3-4 working days

------------------------------------------------------
From: Johannes Weiner <hannes@cmpxchg.org>
Subject: mm: base LRU balancing on an explicit cost model

Currently, scan pressure between the anon and file LRU lists is balanced
based on a mixture of reclaim efficiency and a somewhat vague notion of
"value" of having certain pages in memory over others.  That concept of
value is problematic, because it has caused us to count any event that
remotely makes one LRU list more or less preferrable for reclaim, even
when these events are not directly comparable and impose very different
costs on the system.  One example is referenced file pages that we still
deactivate and referenced anonymous pages that we actually rotate back to
the head of the list.

There is also conceptual overlap with the LRU algorithm itself.  By
rotating recently used pages instead of reclaiming them, the algorithm
already biases the applied scan pressure based on page value.  Thus, when
rebalancing scan pressure due to rotations, we should think of reclaim
cost, and leave assessing the page value to the LRU algorithm.

Lastly, considering both value-increasing as well as value-decreasing
events can sometimes cause the same type of event to be counted twice,
i.e.  how rotating a page increases the LRU value, while reclaiming it
succesfully decreases the value.  In itself this will balance out fine,
but it quietly skews the impact of events that are only recorded once.

The abstract metric of "value", the murky relationship with the LRU
algorithm, and accounting both negative and positive events make the
current pressure balancing model hard to reason about and modify.

This patch switches to a balancing model of accounting the concrete,
actually observed cost of reclaiming one LRU over another.  For now, that
cost includes pages that are scanned but rotated back to the list head. 
Subsequent patches will add consideration for IO caused by refaulting of
recently evicted pages.

Replace struct zone_reclaim_stat with two cost counters in the lruvec, and
make everything that affects cost go through a new lru_note_cost()
function.

Link: http://lkml.kernel.org/r/20200520232525.798933-9-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/mmzone.h |   21 ++++++------------
 include/linux/swap.h   |    2 +
 mm/memcontrol.c        |   18 +++++----------
 mm/swap.c              |   21 +++++++-----------
 mm/vmscan.c            |   44 +++++++++++++++++++--------------------
 5 files changed, 46 insertions(+), 60 deletions(-)

--- a/include/linux/mmzone.h~mm-base-lru-balancing-on-an-explicit-cost-model
+++ a/include/linux/mmzone.h
@@ -240,19 +240,6 @@ static inline bool is_active_lru(enum lr
 	return (lru == LRU_ACTIVE_ANON || lru == LRU_ACTIVE_FILE);
 }
 
-struct zone_reclaim_stat {
-	/*
-	 * The pageout code in vmscan.c keeps track of how many of the
-	 * mem/swap backed and file backed pages are referenced.
-	 * The higher the rotated/scanned ratio, the more valuable
-	 * that cache is.
-	 *
-	 * The anon LRU stats live in [0], file LRU stats in [1]
-	 */
-	unsigned long		recent_rotated[2];
-	unsigned long		recent_scanned[2];
-};
-
 enum lruvec_flags {
 	LRUVEC_CONGESTED,		/* lruvec has many dirty pages
 					 * backed by a congested BDI
@@ -261,7 +248,13 @@ enum lruvec_flags {
 
 struct lruvec {
 	struct list_head		lists[NR_LRU_LISTS];
-	struct zone_reclaim_stat	reclaim_stat;
+	/*
+	 * These track the cost of reclaiming one LRU - file or anon -
+	 * over the other. As the observed cost of reclaiming one LRU
+	 * increases, the reclaim scan balance tips toward the other.
+	 */
+	unsigned long			anon_cost;
+	unsigned long			file_cost;
 	/* Evictions & activations on the inactive file list */
 	atomic_long_t			inactive_age;
 	/* Refaults at the time of last reclaim cycle */
--- a/include/linux/swap.h~mm-base-lru-balancing-on-an-explicit-cost-model
+++ a/include/linux/swap.h
@@ -334,6 +334,8 @@ extern unsigned long nr_free_pagecache_p
 
 
 /* linux/mm/swap.c */
+extern void lru_note_cost(struct lruvec *lruvec, bool file,
+			  unsigned int nr_pages);
 extern void lru_cache_add(struct page *);
 extern void lru_add_page_tail(struct page *page, struct page *page_tail,
 			 struct lruvec *lruvec, struct list_head *head);
--- a/mm/memcontrol.c~mm-base-lru-balancing-on-an-explicit-cost-model
+++ a/mm/memcontrol.c
@@ -3798,23 +3798,17 @@ static int memcg_stat_show(struct seq_fi
 	{
 		pg_data_t *pgdat;
 		struct mem_cgroup_per_node *mz;
-		struct zone_reclaim_stat *rstat;
-		unsigned long recent_rotated[2] = {0, 0};
-		unsigned long recent_scanned[2] = {0, 0};
+		unsigned long anon_cost = 0;
+		unsigned long file_cost = 0;
 
 		for_each_online_pgdat(pgdat) {
 			mz = mem_cgroup_nodeinfo(memcg, pgdat->node_id);
-			rstat = &mz->lruvec.reclaim_stat;
 
-			recent_rotated[0] += rstat->recent_rotated[0];
-			recent_rotated[1] += rstat->recent_rotated[1];
-			recent_scanned[0] += rstat->recent_scanned[0];
-			recent_scanned[1] += rstat->recent_scanned[1];
+			anon_cost += mz->lruvec.anon_cost;
+			file_cost += mz->lruvec.file_cost;
 		}
-		seq_printf(m, "recent_rotated_anon %lu\n", recent_rotated[0]);
-		seq_printf(m, "recent_rotated_file %lu\n", recent_rotated[1]);
-		seq_printf(m, "recent_scanned_anon %lu\n", recent_scanned[0]);
-		seq_printf(m, "recent_scanned_file %lu\n", recent_scanned[1]);
+		seq_printf(m, "anon_cost %lu\n", anon_cost);
+		seq_printf(m, "file_cost %lu\n", file_cost);
 	}
 #endif
 
--- a/mm/swap.c~mm-base-lru-balancing-on-an-explicit-cost-model
+++ a/mm/swap.c
@@ -262,15 +262,12 @@ void rotate_reclaimable_page(struct page
 	}
 }
 
-static void update_page_reclaim_stat(struct lruvec *lruvec,
-				     int file, int rotated,
-				     unsigned int nr_pages)
-{
-	struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
-
-	reclaim_stat->recent_scanned[file] += nr_pages;
-	if (rotated)
-		reclaim_stat->recent_rotated[file] += nr_pages;
+void lru_note_cost(struct lruvec *lruvec, bool file, unsigned int nr_pages)
+{
+	if (file)
+		lruvec->file_cost += nr_pages;
+	else
+		lruvec->anon_cost += nr_pages;
 }
 
 static void __activate_page(struct page *page, struct lruvec *lruvec,
@@ -518,7 +515,7 @@ static void lru_deactivate_file_fn(struc
 
 	if (active)
 		__count_vm_event(PGDEACTIVATE);
-	update_page_reclaim_stat(lruvec, file, 0, hpage_nr_pages(page));
+	lru_note_cost(lruvec, !file, hpage_nr_pages(page));
 }
 
 static void lru_deactivate_fn(struct page *page, struct lruvec *lruvec,
@@ -534,7 +531,7 @@ static void lru_deactivate_fn(struct pag
 		add_page_to_lru_list(page, lruvec, lru);
 
 		__count_vm_events(PGDEACTIVATE, hpage_nr_pages(page));
-		update_page_reclaim_stat(lruvec, file, 0, hpage_nr_pages(page));
+		lru_note_cost(lruvec, !file, hpage_nr_pages(page));
 	}
 }
 
@@ -559,7 +556,7 @@ static void lru_lazyfree_fn(struct page
 
 		__count_vm_events(PGLAZYFREE, hpage_nr_pages(page));
 		count_memcg_page_event(page, PGLAZYFREE);
-		update_page_reclaim_stat(lruvec, 1, 0, hpage_nr_pages(page));
+		lru_note_cost(lruvec, 0, hpage_nr_pages(page));
 	}
 }
 
--- a/mm/vmscan.c~mm-base-lru-balancing-on-an-explicit-cost-model
+++ a/mm/vmscan.c
@@ -1916,7 +1916,6 @@ shrink_inactive_list(unsigned long nr_to
 	bool file = is_file_lru(lru);
 	enum vm_event_item item;
 	struct pglist_data *pgdat = lruvec_pgdat(lruvec);
-	struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
 	bool stalled = false;
 
 	while (unlikely(too_many_isolated(pgdat, file, sc))) {
@@ -1940,7 +1939,6 @@ shrink_inactive_list(unsigned long nr_to
 				     &nr_scanned, sc, lru);
 
 	__mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
-	reclaim_stat->recent_scanned[file] += nr_taken;
 	item = current_is_kswapd() ? PGSCAN_KSWAPD : PGSCAN_DIRECT;
 	if (!cgroup_reclaim(sc))
 		__count_vm_events(item, nr_scanned);
@@ -1960,8 +1958,12 @@ shrink_inactive_list(unsigned long nr_to
 	move_pages_to_lru(lruvec, &page_list);
 
 	__mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, -nr_taken);
-	reclaim_stat->recent_rotated[0] += stat.nr_activate[0];
-	reclaim_stat->recent_rotated[1] += stat.nr_activate[1];
+	/*
+	 * Rotating pages costs CPU without actually
+	 * progressing toward the reclaim goal.
+	 */
+	lru_note_cost(lruvec, 0, stat.nr_activate[0]);
+	lru_note_cost(lruvec, 1, stat.nr_activate[1]);
 	item = current_is_kswapd() ? PGSTEAL_KSWAPD : PGSTEAL_DIRECT;
 	if (!cgroup_reclaim(sc))
 		__count_vm_events(item, nr_reclaimed);
@@ -2013,7 +2015,6 @@ static void shrink_active_list(unsigned
 	LIST_HEAD(l_active);
 	LIST_HEAD(l_inactive);
 	struct page *page;
-	struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
 	unsigned nr_deactivate, nr_activate;
 	unsigned nr_rotated = 0;
 	int file = is_file_lru(lru);
@@ -2027,7 +2028,6 @@ static void shrink_active_list(unsigned
 				     &nr_scanned, sc, lru);
 
 	__mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, nr_taken);
-	reclaim_stat->recent_scanned[file] += nr_taken;
 
 	__count_vm_events(PGREFILL, nr_scanned);
 	__count_memcg_events(lruvec_memcg(lruvec), PGREFILL, nr_scanned);
@@ -2085,7 +2085,7 @@ static void shrink_active_list(unsigned
 	 * helps balance scan pressure between file and anonymous pages in
 	 * get_scan_count.
 	 */
-	reclaim_stat->recent_rotated[file] += nr_rotated;
+	lru_note_cost(lruvec, file, nr_rotated);
 
 	nr_activate = move_pages_to_lru(lruvec, &l_active);
 	nr_deactivate = move_pages_to_lru(lruvec, &l_inactive);
@@ -2242,13 +2242,13 @@ static void get_scan_count(struct lruvec
 {
 	struct mem_cgroup *memcg = lruvec_memcg(lruvec);
 	int swappiness = mem_cgroup_swappiness(memcg);
-	struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
 	u64 fraction[2];
 	u64 denominator = 0;	/* gcc */
 	struct pglist_data *pgdat = lruvec_pgdat(lruvec);
 	unsigned long anon_prio, file_prio;
 	enum scan_balance scan_balance;
 	unsigned long anon, file;
+	unsigned long totalcost;
 	unsigned long ap, fp;
 	enum lru_list lru;
 
@@ -2324,26 +2324,26 @@ static void get_scan_count(struct lruvec
 		lruvec_lru_size(lruvec, LRU_INACTIVE_FILE, MAX_NR_ZONES);
 
 	spin_lock_irq(&pgdat->lru_lock);
-	if (unlikely(reclaim_stat->recent_scanned[0] > anon / 4)) {
-		reclaim_stat->recent_scanned[0] /= 2;
-		reclaim_stat->recent_rotated[0] /= 2;
-	}
-
-	if (unlikely(reclaim_stat->recent_scanned[1] > file / 4)) {
-		reclaim_stat->recent_scanned[1] /= 2;
-		reclaim_stat->recent_rotated[1] /= 2;
+	totalcost = lruvec->anon_cost + lruvec->file_cost;
+	if (unlikely(totalcost > (anon + file) / 4)) {
+		lruvec->anon_cost /= 2;
+		lruvec->file_cost /= 2;
+		totalcost /= 2;
 	}
 
 	/*
 	 * The amount of pressure on anon vs file pages is inversely
-	 * proportional to the fraction of recently scanned pages on
-	 * each list that were recently referenced and in active use.
+	 * proportional to the assumed cost of reclaiming each list,
+	 * as determined by the share of pages that are likely going
+	 * to refault or rotate on each list (recently referenced),
+	 * times the relative IO cost of bringing back a swapped out
+	 * anonymous page vs reloading a filesystem page (swappiness).
 	 */
-	ap = anon_prio * (reclaim_stat->recent_scanned[0] + 1);
-	ap /= reclaim_stat->recent_rotated[0] + 1;
+	ap = anon_prio * (totalcost + 1);
+	ap /= lruvec->anon_cost + 1;
 
-	fp = file_prio * (reclaim_stat->recent_scanned[1] + 1);
-	fp /= reclaim_stat->recent_rotated[1] + 1;
+	fp = file_prio * (totalcost + 1);
+	fp /= lruvec->file_cost + 1;
 	spin_unlock_irq(&pgdat->lru_lock);
 
 	fraction[0] = ap;
_

Patches currently in -mm which might be from hannes@cmpxchg.org are

mm-fix-numa-node-file-count-error-in-replace_page_cache.patch
mm-memcontrol-fix-stat-corrupting-race-in-charge-moving.patch
mm-memcontrol-drop-compound-parameter-from-memcg-charging-api.patch
mm-shmem-remove-rare-optimization-when-swapin-races-with-hole-punching.patch
mm-memcontrol-move-out-cgroup-swaprate-throttling.patch
mm-memcontrol-convert-page-cache-to-a-new-mem_cgroup_charge-api.patch
mm-memcontrol-prepare-uncharging-for-removal-of-private-page-type-counters.patch
mm-memcontrol-prepare-move_account-for-removal-of-private-page-type-counters.patch
mm-memcontrol-prepare-cgroup-vmstat-infrastructure-for-native-anon-counters.patch
mm-memcontrol-switch-to-native-nr_file_pages-and-nr_shmem-counters.patch
mm-memcontrol-switch-to-native-nr_anon_mapped-counter.patch
mm-memcontrol-switch-to-native-nr_anon_thps-counter.patch
mm-memcontrol-switch-to-native-nr_anon_thps-counter-fix.patch
mm-memcontrol-convert-anon-and-file-thp-to-new-mem_cgroup_charge-api.patch
mm-memcontrol-convert-anon-and-file-thp-to-new-mem_cgroup_charge-api-fix.patch
mm-memcontrol-drop-unused-try-commit-cancel-charge-api.patch
mm-memcontrol-prepare-swap-controller-setup-for-integration.patch
mm-memcontrol-make-swap-tracking-an-integral-part-of-memory-control.patch
mm-memcontrol-charge-swapin-pages-on-instantiation.patch
mm-memcontrol-delete-unused-lrucare-handling.patch
mm-memcontrol-update-page-mem_cgroup-stability-rules.patch
mm-fix-lru-balancing-effect-of-new-transparent-huge-pages.patch
mm-keep-separate-anon-and-file-statistics-on-page-reclaim-activity.patch
mm-allow-swappiness-that-prefers-reclaiming-anon-over-the-file-workingset.patch
mm-fold-and-remove-lru_cache_add_anon-and-lru_cache_add_file.patch
mm-workingset-let-cache-workingset-challenge-anon.patch
mm-remove-use-once-cache-bias-from-lru-balancing.patch
mm-vmscan-drop-unnecessary-div0-avoidance-rounding-in-get_scan_count.patch
mm-base-lru-balancing-on-an-explicit-cost-model.patch
mm-deactivations-shouldnt-bias-the-lru-balance.patch
mm-only-count-actual-rotations-as-lru-reclaim-cost.patch
mm-balance-lru-lists-based-on-relative-thrashing.patch
mm-vmscan-determine-anon-file-pressure-balance-at-the-reclaim-root.patch
mm-vmscan-reclaim-writepage-is-io-cost.patch
mm-vmscan-limit-the-range-of-lru-type-balancing.patch