[PATCH 1/2][RESEND] mm: use sc->priority for slab shrink targets

[PATCH 1/2][RESEND] mm: use sc->priority for slab shrink targets

[josef]

From: Josef Bacik <jbacik@fb.com>

Previously we were using the ratio of the number of lru pages scanned to
the number of eligible lru pages to determine the number of slab objects
to scan.  The problem with this is that these two things have nothing to
do with each other, so in slab heavy work loads where there is little to
no page cache we can end up with the pages scanned being a very low
number.  This means that we reclaim next to no slab pages and waste a
lot of time reclaiming small amounts of space.

Instead use sc->priority in the same way we use it to determine scan
amounts for the lru's.  This generally equates to pages.  Consider the
following

slab_pages = (nr_objects * object_size) / PAGE_SIZE

What we would like to do is

scan = slab_pages >> sc->priority

but we don't know the number of slab pages each shrinker controls, only
the objects.  However say that theoretically we knew how many pages a
shrinker controlled, we'd still have to convert this to objects, which
would look like the following

scan = shrinker_pages >> sc->priority
scan_objects = (PAGE_SIZE / object_size) * scan

or written another way

scan_objects = (shrinker_pages >> sc->priority) *
		(PAGE_SIZE / object_size)

which can thus be written

scan_objects = ((shrinker_pages * PAGE_SIZE) / object_size) >>
		sc->priority

which is just

scan_objects = nr_objects >> sc->priority

We don't need to know exactly how many pages each shrinker represents,
it's objects are all the information we need.  Making this change allows
us to place an appropriate amount of pressure on the shrinker pools for
their relative size.

Signed-off-by: Josef Bacik <jbacik@fb.com>
---
 include/trace/events/vmscan.h | 23 ++++++++++------------
 mm/vmscan.c                   | 46 +++++++++++--------------------------------
 2 files changed, 22 insertions(+), 47 deletions(-)

diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
index 27e8a5c..8c5a00a 100644
--- a/include/trace/events/vmscan.h
+++ b/include/trace/events/vmscan.h
@@ -187,12 +187,12 @@ DEFINE_EVENT(mm_vmscan_direct_reclaim_end_template, mm_vmscan_memcg_softlimit_re
 
 TRACE_EVENT(mm_shrink_slab_start,
 	TP_PROTO(struct shrinker *shr, struct shrink_control *sc,
-		long nr_objects_to_shrink, unsigned long pgs_scanned,
-		unsigned long lru_pgs, unsigned long cache_items,
-		unsigned long long delta, unsigned long total_scan),
+		long nr_objects_to_shrink, unsigned long cache_items,
+		unsigned long long delta, unsigned long total_scan,
+		int priority),
 
-	TP_ARGS(shr, sc, nr_objects_to_shrink, pgs_scanned, lru_pgs,
-		cache_items, delta, total_scan),
+	TP_ARGS(shr, sc, nr_objects_to_shrink, cache_items, delta, total_scan,
+		priority),
 
 	TP_STRUCT__entry(
 		__field(struct shrinker *, shr)
@@ -200,11 +200,10 @@ TRACE_EVENT(mm_shrink_slab_start,
 		__field(int, nid)
 		__field(long, nr_objects_to_shrink)
 		__field(gfp_t, gfp_flags)
-		__field(unsigned long, pgs_scanned)
-		__field(unsigned long, lru_pgs)
 		__field(unsigned long, cache_items)
 		__field(unsigned long long, delta)
 		__field(unsigned long, total_scan)
+		__field(int, priority)
 	),
 
 	TP_fast_assign(
@@ -213,24 +212,22 @@ TRACE_EVENT(mm_shrink_slab_start,
 		__entry->nid = sc->nid;
 		__entry->nr_objects_to_shrink = nr_objects_to_shrink;
 		__entry->gfp_flags = sc->gfp_mask;
-		__entry->pgs_scanned = pgs_scanned;
-		__entry->lru_pgs = lru_pgs;
 		__entry->cache_items = cache_items;
 		__entry->delta = delta;
 		__entry->total_scan = total_scan;
+		__entry->priority = priority;
 	),
 
-	TP_printk("%pF %p: nid: %d objects to shrink %ld gfp_flags %s pgs_scanned %ld lru_pgs %ld cache items %ld delta %lld total_scan %ld",
+	TP_printk("%pF %p: nid: %d objects to shrink %ld gfp_flags %s cache items %ld delta %lld total_scan %ld priority %d",
 		__entry->shrink,
 		__entry->shr,
 		__entry->nid,
 		__entry->nr_objects_to_shrink,
 		show_gfp_flags(__entry->gfp_flags),
-		__entry->pgs_scanned,
-		__entry->lru_pgs,
 		__entry->cache_items,
 		__entry->delta,
-		__entry->total_scan)
+		__entry->total_scan,
+		__entry->priority)
 );
 
 TRACE_EVENT(mm_shrink_slab_end,
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 734e8d3..608dfe6 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -306,9 +306,7 @@ EXPORT_SYMBOL(unregister_shrinker);
 #define SHRINK_BATCH 128
 
 static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
-				    struct shrinker *shrinker,
-				    unsigned long nr_scanned,
-				    unsigned long nr_eligible)
+				    struct shrinker *shrinker, int priority)
 {
 	unsigned long freed = 0;
 	unsigned long long delta;
@@ -333,9 +331,8 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
 	nr = atomic_long_xchg(&shrinker->nr_deferred[nid], 0);
 
 	total_scan = nr;
-	delta = (4 * nr_scanned) / shrinker->seeks;
-	delta *= freeable;
-	do_div(delta, nr_eligible + 1);
+	delta = freeable >> priority;
+	delta = (4 * freeable) / shrinker->seeks;
 	total_scan += delta;
 	if (total_scan < 0) {
 		pr_err("shrink_slab: %pF negative objects to delete nr=%ld\n",
@@ -369,8 +366,7 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
 		total_scan = freeable * 2;
 
 	trace_mm_shrink_slab_start(shrinker, shrinkctl, nr,
-				   nr_scanned, nr_eligible,
-				   freeable, delta, total_scan);
+				   freeable, delta, total_scan, priority);
 
 	/*
 	 * Normally, we should not scan less than batch_size objects in one
@@ -429,8 +425,7 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
  * @gfp_mask: allocation context
  * @nid: node whose slab caches to target
  * @memcg: memory cgroup whose slab caches to target
- * @nr_scanned: pressure numerator
- * @nr_eligible: pressure denominator
+ * @priority: the reclaim priority
  *
  * Call the shrink functions to age shrinkable caches.
  *
@@ -442,20 +437,14 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
  * objects from the memory cgroup specified. Otherwise, only unaware
  * shrinkers are called.
  *
- * @nr_scanned and @nr_eligible form a ratio that indicate how much of
- * the available objects should be scanned.  Page reclaim for example
- * passes the number of pages scanned and the number of pages on the
- * LRU lists that it considered on @nid, plus a bias in @nr_scanned
- * when it encountered mapped pages.  The ratio is further biased by
- * the ->seeks setting of the shrink function, which indicates the
- * cost to recreate an object relative to that of an LRU page.
+ * @priority is sc->priority, we take the number of objects and >> by priority
+ * in order to get the scan target.
  *
  * Returns the number of reclaimed slab objects.
  */
 static unsigned long shrink_slab(gfp_t gfp_mask, int nid,
 				 struct mem_cgroup *memcg,
-				 unsigned long nr_scanned,
-				 unsigned long nr_eligible)
+				 int priority)
 {
 	struct shrinker *shrinker;
 	unsigned long freed = 0;
@@ -463,9 +452,6 @@ static unsigned long shrink_slab(gfp_t gfp_mask, int nid,
 	if (memcg && (!memcg_kmem_enabled() || !mem_cgroup_online(memcg)))
 		return 0;
 
-	if (nr_scanned == 0)
-		nr_scanned = SWAP_CLUSTER_MAX;
-
 	if (!down_read_trylock(&shrinker_rwsem)) {
 		/*
 		 * If we would return 0, our callers would understand that we
@@ -496,7 +482,7 @@ static unsigned long shrink_slab(gfp_t gfp_mask, int nid,
 		if (!(shrinker->flags & SHRINKER_NUMA_AWARE))
 			sc.nid = 0;
 
-		freed += do_shrink_slab(&sc, shrinker, nr_scanned, nr_eligible);
+		freed += do_shrink_slab(&sc, shrinker, priority);
 	}
 
 	up_read(&shrinker_rwsem);
@@ -514,8 +500,7 @@ void drop_slab_node(int nid)
 
 		freed = 0;
 		do {
-			freed += shrink_slab(GFP_KERNEL, nid, memcg,
-					     1000, 1000);
+			freed += shrink_slab(GFP_KERNEL, nid, memcg, 0);
 		} while ((memcg = mem_cgroup_iter(NULL, memcg, NULL)) != NULL);
 	} while (freed > 10);
 }
@@ -2610,14 +2595,12 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 
 			reclaimed = sc->nr_reclaimed;
 			scanned = sc->nr_scanned;
-
 			shrink_node_memcg(pgdat, memcg, sc, &lru_pages);
 			node_lru_pages += lru_pages;
 
 			if (memcg)
 				shrink_slab(sc->gfp_mask, pgdat->node_id,
-					    memcg, sc->nr_scanned - scanned,
-					    lru_pages);
+					    memcg, sc->priority);
 
 			/* Record the group's reclaim efficiency */
 			vmpressure(sc->gfp_mask, memcg, false,
@@ -2641,14 +2624,9 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 			}
 		} while ((memcg = mem_cgroup_iter(root, memcg, &reclaim)));
 
-		/*
-		 * Shrink the slab caches in the same proportion that
-		 * the eligible LRU pages were scanned.
-		 */
 		if (global_reclaim(sc))
 			shrink_slab(sc->gfp_mask, pgdat->node_id, NULL,
-				    sc->nr_scanned - nr_scanned,
-				    node_lru_pages);
+				    sc->priority);
 
 		/*
 		 * Record the subtree's reclaim efficiency. The reclaimed
-- 
2.7.4

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH 2/2][RESEND] mm: make kswapd try harder to keep active pages in cache

[josef]

From: Josef Bacik <jbacik@fb.com>

While testing slab reclaim I noticed that if we were running a workload
that used most of the system memory for it's working set and we start
putting a lot of reclaimable slab pressure on the system (think find /,
or some other silliness), we will happily evict the active pages over
the slab cache.  This is kind of backwards as we want to do all that we
can to keep the active working set in memory, and instead evict these
short lived objects.  The same thing occurs when say you do a yum
update of a few packages while your working set takes up most of RAM,
you end up with inactive lists being relatively small and so we reclaim
active pages even though we could reclaim these short lived inactive
pages.

My approach here is twofold.  First, keep track of the difference in
inactive and slab pages since the last time kswapd ran.  In the first
run this will just be the overall counts of inactive and slab, but for
each subsequent run we'll have a good idea of where the memory pressure
is coming from.  Then we use this information to put pressure on either
the inactive lists or the slab caches, depending on where the pressure
is coming from.

I have two tests I was using to watch either side of this problem.  The
first test kept 2 files that took up 3/4 of the memory, and then started
creating a bunch of empty files.  Without this patch we would have to
re-read both files in their entirety at least 3 times during the run.
With this patch the active pages are never evicted.

The second test was a test that would read and stat all the files in a
directory, which again would take up about 3/4 of the memory with slab
cache.  Then I cat'ed a 100gib file into /dev/null and checked to see if
any of the files were evicted and verified that none of the files were
evicted.

Signed-off-by: Josef Bacik <jbacik@fb.com>
---
 mm/vmscan.c | 172 +++++++++++++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 158 insertions(+), 14 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 608dfe6..d6ba41b 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -110,11 +110,20 @@ struct scan_control {
 	/* One of the zones is ready for compaction */
 	unsigned int compaction_ready:1;
 
+	/* Only reclaim inactive page cache or slab. */
+	unsigned int inactive_only:1;
+
 	/* Incremented by the number of inactive pages that were scanned */
 	unsigned long nr_scanned;
 
 	/* Number of pages freed so far during a call to shrink_zones() */
 	unsigned long nr_reclaimed;
+
+	/* Number of inactive pages added since last kswapd run. */
+	unsigned long inactive_diff;
+
+	/* Number of slab pages added since last kswapd run. */
+	unsigned long slab_diff;
 };
 
 #ifdef ARCH_HAS_PREFETCH
@@ -306,7 +315,8 @@ EXPORT_SYMBOL(unregister_shrinker);
 #define SHRINK_BATCH 128
 
 static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
-				    struct shrinker *shrinker, int priority)
+				    struct shrinker *shrinker, int priority,
+				    unsigned long *slab_scanned)
 {
 	unsigned long freed = 0;
 	unsigned long long delta;
@@ -405,6 +415,9 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
 		next_deferred -= scanned;
 	else
 		next_deferred = 0;
+	if (slab_scanned)
+		(*slab_scanned) += scanned;
+
 	/*
 	 * move the unused scan count back into the shrinker in a
 	 * manner that handles concurrent updates. If we exhausted the
@@ -444,7 +457,7 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
  */
 static unsigned long shrink_slab(gfp_t gfp_mask, int nid,
 				 struct mem_cgroup *memcg,
-				 int priority)
+				 int priority, unsigned long *slab_scanned)
 {
 	struct shrinker *shrinker;
 	unsigned long freed = 0;
@@ -482,7 +495,7 @@ static unsigned long shrink_slab(gfp_t gfp_mask, int nid,
 		if (!(shrinker->flags & SHRINKER_NUMA_AWARE))
 			sc.nid = 0;
 
-		freed += do_shrink_slab(&sc, shrinker, priority);
+		freed += do_shrink_slab(&sc, shrinker, priority, slab_scanned);
 	}
 
 	up_read(&shrinker_rwsem);
@@ -500,7 +513,7 @@ void drop_slab_node(int nid)
 
 		freed = 0;
 		do {
-			freed += shrink_slab(GFP_KERNEL, nid, memcg, 0);
+			freed += shrink_slab(GFP_KERNEL, nid, memcg, 0, NULL);
 		} while ((memcg = mem_cgroup_iter(NULL, memcg, NULL)) != NULL);
 	} while (freed > 10);
 }
@@ -2149,6 +2162,7 @@ enum scan_balance {
 	SCAN_FRACT,
 	SCAN_ANON,
 	SCAN_FILE,
+	SCAN_INACTIVE,
 };
 
 /*
@@ -2175,6 +2189,11 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg,
 	unsigned long ap, fp;
 	enum lru_list lru;
 
+	if (sc->inactive_only) {
+		scan_balance = SCAN_INACTIVE;
+		goto out;
+	}
+
 	/* If we have no swap space, do not bother scanning anon pages. */
 	if (!sc->may_swap || mem_cgroup_get_nr_swap_pages(memcg) <= 0) {
 		scan_balance = SCAN_FILE;
@@ -2348,6 +2367,14 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg,
 				scan = 0;
 			}
 			break;
+		case SCAN_INACTIVE:
+			if (file && !is_active_lru(lru)) {
+				scan = max(scan, sc->nr_to_reclaim);
+			} else {
+				size = 0;
+				scan = 0;
+			}
+			break;
 		default:
 			/* Look ma, no brain */
 			BUG();
@@ -2565,7 +2592,61 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 {
 	struct reclaim_state *reclaim_state = current->reclaim_state;
 	unsigned long nr_reclaimed, nr_scanned;
+	unsigned long greclaim = 1, gslab = 1, total_high_wmark = 0, nr_inactive;
+	int priority_adj = 1;
 	bool reclaimable = false;
+	bool skip_slab = false;
+
+	if (global_reclaim(sc)) {
+		int z;
+		for (z = 0; z < MAX_NR_ZONES; z++) {
+			struct zone *zone = &pgdat->node_zones[z];
+			if (!managed_zone(zone))
+				continue;
+			total_high_wmark += high_wmark_pages(zone);
+		}
+		nr_inactive = node_page_state(pgdat, NR_INACTIVE_FILE);
+		gslab = node_page_state(pgdat, NR_SLAB_RECLAIMABLE);
+		greclaim = pgdat_reclaimable_pages(pgdat);
+	} else {
+		struct lruvec *lruvec =
+			mem_cgroup_lruvec(pgdat, sc->target_mem_cgroup);
+		total_high_wmark = sc->nr_to_reclaim;
+		nr_inactive = lruvec_page_state(lruvec, NR_INACTIVE_FILE);
+		gslab = lruvec_page_state(lruvec, NR_SLAB_RECLAIMABLE);
+	}
+
+	/*
+	 * If we don't have a lot of inactive or slab pages then there's no
+	 * point in trying to free them exclusively, do the normal scan stuff.
+	 */
+	if (nr_inactive + gslab < total_high_wmark)
+		sc->inactive_only = 0;
+
+	/*
+	 * We still want to slightly prefer slab over inactive, so if the
+	 * inactive on this node is large enough and what is pushing us into
+	 * reclaim territory then limit our flushing to the inactive list for
+	 * the first go around.
+	 *
+	 * The idea is that with a memcg configured system we will still reclaim
+	 * memcg aware shrinkers, which includes the super block shrinkers.  So
+	 * if our steady state is keeping fs objects in cache for our workload
+	 * we'll still put a certain amount of pressure on them anyway.  To
+	 * avoid evicting things we actually care about we want to skip slab
+	 * reclaim altogether.
+	 *
+	 * However we still want to account for slab and inactive growing at the
+	 * same rate, so if that is the case just carry on shrinking inactive
+	 * and slab together.
+	 */
+	if (nr_inactive > total_high_wmark &&
+	    sc->inactive_diff > sc->slab_diff) {
+		unsigned long tmp = sc->inactive_diff >> 1;
+
+		if (tmp >= sc->slab_diff)
+			skip_slab = true;
+	}
 
 	do {
 		struct mem_cgroup *root = sc->target_mem_cgroup;
@@ -2574,6 +2655,8 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 			.priority = sc->priority,
 		};
 		unsigned long node_lru_pages = 0;
+		unsigned long slab_reclaimed = 0;
+		unsigned long slab_scanned = 0;
 		struct mem_cgroup *memcg;
 
 		nr_reclaimed = sc->nr_reclaimed;
@@ -2598,9 +2681,16 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 			shrink_node_memcg(pgdat, memcg, sc, &lru_pages);
 			node_lru_pages += lru_pages;
 
-			if (memcg)
-				shrink_slab(sc->gfp_mask, pgdat->node_id,
-					    memcg, sc->priority);
+			if (memcg && !skip_slab) {
+				int priority = sc->priority;
+				if (sc->inactive_only)
+					priority -= priority_adj;
+				priority = max(0, priority);
+				slab_reclaimed +=
+					shrink_slab(sc->gfp_mask,
+						    pgdat->node_id, memcg,
+						    priority, &slab_scanned);
+			}
 
 			/* Record the group's reclaim efficiency */
 			vmpressure(sc->gfp_mask, memcg, false,
@@ -2624,9 +2714,16 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 			}
 		} while ((memcg = mem_cgroup_iter(root, memcg, &reclaim)));
 
-		if (global_reclaim(sc))
-			shrink_slab(sc->gfp_mask, pgdat->node_id, NULL,
-				    sc->priority);
+		if (!skip_slab && global_reclaim(sc)) {
+			int priority = sc->priority;
+			if (sc->inactive_only)
+				priority -= priority_adj;
+			priority = max(0, priority);
+			slab_reclaimed +=
+				shrink_slab(sc->gfp_mask, pgdat->node_id, NULL,
+					    priority, &slab_scanned);
+		}
+
 
 		/*
 		 * Record the subtree's reclaim efficiency. The reclaimed
@@ -2645,9 +2742,28 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
 			reclaim_state->reclaimed_slab = 0;
 		}
 
-		if (sc->nr_reclaimed - nr_reclaimed)
+		if (sc->nr_reclaimed - nr_reclaimed) {
 			reclaimable = true;
+		} else if (sc->inactive_only && !skip_slab) {
+			unsigned long percent = 0;
 
+			/*
+			 * We didn't reclaim anything this go around, so the
+			 * inactive list is likely spent.  If we're reclaiming
+			 * less than half of the objects in slab that we're
+			 * scanning then just stop doing the inactive only scan.
+			 * Otherwise ramp up the pressure on the slab caches
+			 * hoping that eventually we'll start freeing enough
+			 * objects to reclaim space.
+			 */
+			if (slab_scanned)
+				percent = slab_reclaimed * 100 / slab_scanned;
+			if (percent < 50)
+				sc->inactive_only = 0;
+			else
+				priority_adj++;
+		}
+		skip_slab = false;
 	} while (should_continue_reclaim(pgdat, sc->nr_reclaimed - nr_reclaimed,
 					 sc->nr_scanned - nr_scanned, sc));
 
@@ -3290,7 +3406,8 @@ static bool kswapd_shrink_node(pg_data_t *pgdat,
  * or lower is eligible for reclaim until at least one usable zone is
  * balanced.
  */
-static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
+static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx,
+			 unsigned long inactive_diff, unsigned long slab_diff)
 {
 	int i;
 	unsigned long nr_soft_reclaimed;
@@ -3303,6 +3420,9 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
 		.may_writepage = !laptop_mode,
 		.may_unmap = 1,
 		.may_swap = 1,
+		.inactive_only = 1,
+		.inactive_diff = inactive_diff,
+		.slab_diff = slab_diff,
 	};
 	count_vm_event(PAGEOUTRUN);
 
@@ -3522,7 +3642,7 @@ static int kswapd(void *p)
 	unsigned int classzone_idx = MAX_NR_ZONES - 1;
 	pg_data_t *pgdat = (pg_data_t*)p;
 	struct task_struct *tsk = current;
-
+	unsigned long nr_slab = 0, nr_inactive = 0;
 	struct reclaim_state reclaim_state = {
 		.reclaimed_slab = 0,
 	};
@@ -3552,6 +3672,7 @@ static int kswapd(void *p)
 	pgdat->kswapd_order = 0;
 	pgdat->kswapd_classzone_idx = MAX_NR_ZONES;
 	for ( ; ; ) {
+		unsigned long slab_diff, inactive_diff;
 		bool ret;
 
 		alloc_order = reclaim_order = pgdat->kswapd_order;
@@ -3579,6 +3700,23 @@ static int kswapd(void *p)
 			continue;
 
 		/*
+		 * We want to know where we're adding pages so we can make
+		 * smarter decisions about where we're going to put pressure
+		 * when shrinking.
+		 */
+		slab_diff = sum_zone_node_page_state(pgdat->node_id,
+						     NR_SLAB_RECLAIMABLE);
+		inactive_diff = node_page_state(pgdat, NR_INACTIVE_FILE);
+		if (nr_slab > slab_diff)
+			slab_diff = 0;
+		else
+			slab_diff -= nr_slab;
+		if (inactive_diff < nr_inactive)
+			inactive_diff = 0;
+		else
+			inactive_diff -= nr_inactive;
+
+		/*
 		 * Reclaim begins at the requested order but if a high-order
 		 * reclaim fails then kswapd falls back to reclaiming for
 		 * order-0. If that happens, kswapd will consider sleeping
@@ -3588,7 +3726,11 @@ static int kswapd(void *p)
 		 */
 		trace_mm_vmscan_kswapd_wake(pgdat->node_id, classzone_idx,
 						alloc_order);
-		reclaim_order = balance_pgdat(pgdat, alloc_order, classzone_idx);
+		reclaim_order = balance_pgdat(pgdat, alloc_order, classzone_idx,
+					      inactive_diff, slab_diff);
+		nr_inactive = node_page_state(pgdat, NR_INACTIVE_FILE);
+		nr_slab = sum_zone_node_page_state(pgdat->node_id,
+						   NR_SLAB_RECLAIMABLE);
 		if (reclaim_order < alloc_order)
 			goto kswapd_try_sleep;
 	}
@@ -3840,6 +3982,8 @@ static int __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned in
 		.may_unmap = !!(node_reclaim_mode & RECLAIM_UNMAP),
 		.may_swap = 1,
 		.reclaim_idx = gfp_zone(gfp_mask),
+		.slab_diff = 1,
+		.inactive_diff = 1,
 	};
 
 	cond_resched();
-- 
2.7.4

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 2/2][RESEND] mm: make kswapd try harder to keep active pages in cache

[Andrey Ryabinin]

On 08/18/2017 05:28 PM, josef@toxicpanda.com wrote:

> @@ -3552,6 +3672,7 @@ static int kswapd(void *p)
>  	pgdat->kswapd_order = 0;
>  	pgdat->kswapd_classzone_idx = MAX_NR_ZONES;
>  	for ( ; ; ) {
> +		unsigned long slab_diff, inactive_diff;
>  		bool ret;
>  
>  		alloc_order = reclaim_order = pgdat->kswapd_order;
> @@ -3579,6 +3700,23 @@ static int kswapd(void *p)
>  			continue;
>  
>  		/*
> +		 * We want to know where we're adding pages so we can make
> +		 * smarter decisions about where we're going to put pressure
> +		 * when shrinking.
> +		 */
> +		slab_diff = sum_zone_node_page_state(pgdat->node_id,
> +						     NR_SLAB_RECLAIMABLE);

This should be node_page_state().

sum_zone_node_page_state() counts sum of zone_stat_item counter across node, but
NR_SLAB_RECLAIMABLE is node_stat_item counter. So in fact you'll get NR_ZONE_UNEVICTABLE here,
(since value of NR_ZONE_UNEVICTABLE equals to NR_SLAB_RECLAIMABLE).


> +		inactive_diff = node_page_state(pgdat, NR_INACTIVE_FILE);
> +		if (nr_slab > slab_diff)
> +			slab_diff = 0;
> +		else
> +			slab_diff -= nr_slab;
> +		if (inactive_diff < nr_inactive)
> +			inactive_diff = 0;
> +		else
> +			inactive_diff -= nr_inactive;
> +
> +		/*
>  		 * Reclaim begins at the requested order but if a high-order
>  		 * reclaim fails then kswapd falls back to reclaiming for
>  		 * order-0. If that happens, kswapd will consider sleeping
> @@ -3588,7 +3726,11 @@ static int kswapd(void *p)
>  		 */
>  		trace_mm_vmscan_kswapd_wake(pgdat->node_id, classzone_idx,
>  						alloc_order);
> -		reclaim_order = balance_pgdat(pgdat, alloc_order, classzone_idx);
> +		reclaim_order = balance_pgdat(pgdat, alloc_order, classzone_idx,
> +					      inactive_diff, slab_diff);
> +		nr_inactive = node_page_state(pgdat, NR_INACTIVE_FILE);
> +		nr_slab = sum_zone_node_page_state(pgdat->node_id,
> +						   NR_SLAB_RECLAIMABLE);

ditto.

>  		if (reclaim_order < alloc_order)
>  			goto kswapd_try_sleep;
>  	}
> @@ -3840,6 +3982,8 @@ static int __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned in
>  		.may_unmap = !!(node_reclaim_mode & RECLAIM_UNMAP),
>  		.may_swap = 1,
>  		.reclaim_idx = gfp_zone(gfp_mask),
> +		.slab_diff = 1,
> +		.inactive_diff = 1,

The only place where __node_reclaim() may use these fields is in the shrink_node():

		if (nr_inactive > total_high_wmark &&
		    sc->inactive_diff > sc->slab_diff) {

Obviously 1 vs 0 doesn't make any difference here. This makes we wonder, why are these fields initialized to 1?


>  	};
>  
>  	cond_resched();
> 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>