6.5: mm patches

lib/scatterlist.c

@@ -150,31 +150,12 @@ EXPORT_SYMBOL(sg_init_one);
  */
 static struct scatterlist *sg_kmalloc(unsigned int nents, gfp_t gfp_mask)
 {
-	if (nents == SG_MAX_SINGLE_ALLOC) {
-		/*
-		 * Kmemleak doesn't track page allocations as they are not
-		 * commonly used (in a raw form) for kernel data structures.
-		 * As we chain together a list of pages and then a normal
-		 * kmalloc (tracked by kmemleak), in order to for that last
-		 * allocation not to become decoupled (and thus a
-		 * false-positive) we need to inform kmemleak of all the
-		 * intermediate allocations.
-		 */
-		void *ptr = (void *) __get_free_page(gfp_mask);
-		kmemleak_alloc(ptr, PAGE_SIZE, 1, gfp_mask);
-		return ptr;
-	} else
-		return kmalloc_array(nents, sizeof(struct scatterlist),
-				     gfp_mask);
+	return kmalloc_array(nents, sizeof(struct scatterlist), gfp_mask);
 }
 
 static void sg_kfree(struct scatterlist *sg, unsigned int nents)
 {
-	if (nents == SG_MAX_SINGLE_ALLOC) {
-		kmemleak_free(sg);
-		free_page((unsigned long) sg);
-	} else
-		kfree(sg);
+	kfree(sg);
 }
 
 /**

mm/compaction.c

@@ -1780,7 +1780,7 @@ static int sysctl_compact_unevictable_allowed __read_mostly = CONFIG_COMPACT_UNE
  * aggressively the kernel should compact memory in the
  * background. It takes values in the range [0, 100].
  */
-static unsigned int __read_mostly sysctl_compaction_proactiveness = 20;
+static unsigned int __read_mostly sysctl_compaction_proactiveness;
 static int sysctl_extfrag_threshold = 500;
 static int __read_mostly sysctl_compact_memory;
 

mm/internal.h

@@ -421,6 +421,7 @@ extern void prep_compound_page(struct page *page, unsigned int order);
 extern void post_alloc_hook(struct page *page, unsigned int order,
 					gfp_t gfp_flags);
 extern int user_min_free_kbytes;
+extern atomic_long_t kswapd_waiters;
 
 extern void free_unref_page(struct page *page, unsigned int order);
 extern void free_unref_page_list(struct list_head *list);

mm/list_lru.c

@@ -178,6 +178,7 @@ EXPORT_SYMBOL_GPL(list_lru_isolate_move);
 unsigned long list_lru_count_one(struct list_lru *lru,
 				 int nid, struct mem_cgroup *memcg)
 {
+#if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB)
 	struct list_lru_one *l;
 	long count;
 
@@ -190,6 +191,9 @@ unsigned long list_lru_count_one(struct list_lru *lru,
 		count = 0;
 
 	return count;
+#else
+	return READ_ONCE(lru->node[nid].lru.nr_items);
+#endif
 }
 EXPORT_SYMBOL_GPL(list_lru_count_one);
 

mm/page_alloc.c

@@ -204,6 +204,8 @@ EXPORT_SYMBOL(node_states);
 
 gfp_t gfp_allowed_mask __read_mostly = GFP_BOOT_MASK;
 
+atomic_long_t kswapd_waiters = ATOMIC_LONG_INIT(0);
+
 /*
  * A cached value of the page's pageblock's migratetype, used when the page is
  * put on a pcplist. Used to avoid the pageblock migratetype lookup when
@@ -297,7 +299,7 @@ static compound_page_dtor * const compound_page_dtors[NR_COMPOUND_DTORS] = {
 
 int min_free_kbytes = 1024;
 int user_min_free_kbytes = -1;
-static int watermark_boost_factor __read_mostly = 15000;
+static int watermark_boost_factor __read_mostly;
 static int watermark_scale_factor = 10;
 
 /* movable_zone is the "real" zone pages in ZONE_MOVABLE are taken from */
@@ -2152,16 +2154,17 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
 }
 
 /*
- * Obtain a specified number of elements from the buddy allocator, all under
- * a single hold of the lock, for efficiency.  Add them to the supplied list.
- * Returns the number of new pages which were placed at *list.
+ * Obtain a specified number of elements from the buddy allocator, and relax the
+ * zone lock when needed. Add them to the supplied list. Returns the number of
+ * new pages which were placed at *list.
  */
 static int rmqueue_bulk(struct zone *zone, unsigned int order,
 			unsigned long count, struct list_head *list,
 			int migratetype, unsigned int alloc_flags)
 {
+	const bool can_resched = !preempt_count() && !irqs_disabled();
 	unsigned long flags;
-	int i;
+	int i, last_mod = 0;
 
 	spin_lock_irqsave(&zone->lock, flags);
 	for (i = 0; i < count; ++i) {
@@ -2170,6 +2173,18 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
 		if (unlikely(page == NULL))
 			break;
 
+		/* Reschedule and ease the contention on the lock if needed */
+		if (i + 1 < count && ((can_resched && need_resched()) ||
+				      spin_needbreak(&zone->lock))) {
+			__mod_zone_page_state(zone, NR_FREE_PAGES,
+					      -((i + 1 - last_mod) << order));
+			last_mod = i + 1;
+			spin_unlock_irqrestore(&zone->lock, flags);
+			if (can_resched)
+				cond_resched();
+			spin_lock_irqsave(&zone->lock, flags);
+		}
+
 		/*
 		 * Split buddy pages returned by expand() are received here in
 		 * physical page order. The page is added to the tail of
@@ -2186,7 +2201,7 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
 					      -(1 << order));
 	}
 
-	__mod_zone_page_state(zone, NR_FREE_PAGES, -(i << order));
+	__mod_zone_page_state(zone, NR_FREE_PAGES, -((i - last_mod) << order));
 	spin_unlock_irqrestore(&zone->lock, flags);
 
 	return i;
@@ -3962,6 +3977,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	unsigned int cpuset_mems_cookie;
 	unsigned int zonelist_iter_cookie;
 	int reserve_flags;
+	bool woke_kswapd = false;
 
 restart:
 	compaction_retries = 0;
@@ -4001,8 +4017,13 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 			goto nopage;
 	}
 
-	if (alloc_flags & ALLOC_KSWAPD)
+	if (alloc_flags & ALLOC_KSWAPD) {
+		if (!woke_kswapd) {
+			atomic_long_inc(&kswapd_waiters);
+			woke_kswapd = true;
+		}
 		wake_all_kswapds(order, gfp_mask, ac);
+	}
 
 	/*
 	 * The adjusted alloc_flags might result in immediate success, so try
@@ -4217,9 +4238,12 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 		goto retry;
 	}
 fail:
-	warn_alloc(gfp_mask, ac->nodemask,
-			"page allocation failure: order:%u", order);
 got_pg:
+	if (woke_kswapd)
+		atomic_long_dec(&kswapd_waiters);
+	if (!page)
+		warn_alloc(gfp_mask, ac->nodemask,
+				"page allocation failure: order:%u", order);
 	return page;
 }
 

mm/vmscan.c

@@ -6901,7 +6901,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
 	return 0;
 }
 
-static bool allow_direct_reclaim(pg_data_t *pgdat)
+static bool allow_direct_reclaim(pg_data_t *pgdat, bool using_kswapd)
 {
 	struct zone *zone;
 	unsigned long pfmemalloc_reserve = 0;
@@ -6930,6 +6930,10 @@ static bool allow_direct_reclaim(pg_data_t *pgdat)
 
 	wmark_ok = free_pages > pfmemalloc_reserve / 2;
 
+	/* The throttled direct reclaimer is now a kswapd waiter */
+	if (unlikely(!using_kswapd && !wmark_ok))
+		atomic_long_inc(&kswapd_waiters);
+
 	/* kswapd must be awake if processes are being throttled */
 	if (!wmark_ok && waitqueue_active(&pgdat->kswapd_wait)) {
 		if (READ_ONCE(pgdat->kswapd_highest_zoneidx) > ZONE_NORMAL)
@@ -6995,7 +6999,7 @@ static bool throttle_direct_reclaim(gfp_t gfp_mask, struct zonelist *zonelist,
 
 		/* Throttle based on the first usable node */
 		pgdat = zone->zone_pgdat;
-		if (allow_direct_reclaim(pgdat))
+		if (allow_direct_reclaim(pgdat, gfp_mask & __GFP_KSWAPD_RECLAIM))
 			goto out;
 		break;
 	}
@@ -7017,11 +7021,14 @@ static bool throttle_direct_reclaim(gfp_t gfp_mask, struct zonelist *zonelist,
 	 */
 	if (!(gfp_mask & __GFP_FS))
 		wait_event_interruptible_timeout(pgdat->pfmemalloc_wait,
-			allow_direct_reclaim(pgdat), HZ);
+			allow_direct_reclaim(pgdat, true), HZ);
 	else
 		/* Throttle until kswapd wakes the process */
 		wait_event_killable(zone->zone_pgdat->pfmemalloc_wait,
-			allow_direct_reclaim(pgdat));
+			allow_direct_reclaim(pgdat, true));
+
+	if (unlikely(!(gfp_mask & __GFP_KSWAPD_RECLAIM)))
+		atomic_long_dec(&kswapd_waiters);
 
 	if (fatal_signal_pending(current))
 		return true;
@@ -7519,14 +7526,15 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int highest_zoneidx)
 		 * able to safely make forward progress. Wake them
 		 */
 		if (waitqueue_active(&pgdat->pfmemalloc_wait) &&
-				allow_direct_reclaim(pgdat))
+				allow_direct_reclaim(pgdat, true))
 			wake_up_all(&pgdat->pfmemalloc_wait);
 
 		/* Check if kswapd should be suspending */
 		__fs_reclaim_release(_THIS_IP_);
 		ret = try_to_freeze();
 		__fs_reclaim_acquire(_THIS_IP_);
-		if (ret || kthread_should_stop())
+		if (ret || kthread_should_stop() ||
+		    !atomic_long_read(&kswapd_waiters))
 			break;
 
 		/*