Merge "vmscan: Support multiple kswapd threads per node"

Documentation/admin-guide/sysctl/vm.rst

@@ -39,6 +39,7 @@ Currently, these files are in /proc/sys/vm:
 - extfrag_threshold
 - extra_free_kbytes
 - hugetlb_shm_group
+- kswapd_threads
 - laptop_mode
 - legacy_va_layout
 - lowmem_reserve_ratio
@@ -310,6 +311,25 @@ hugetlb_shm_group
 hugetlb_shm_group contains group id that is allowed to create SysV
 shared memory segment using hugetlb page.
 
+kswapd_threads
+==============
+kswapd_threads allows you to control the number of kswapd threads per node
+running on the system. This provides the ability to devote additional CPU
+resources toward proactive page replacement with the goal of reducing
+direct reclaims. When direct reclaims are prevented, the CPU consumed
+by them is prevented as well. Depending on the workload, the result can
+cause aggregate CPU usage on the system to go up, down or stay the same.
+
+More aggressive page replacement can reduce direct reclaims which cause
+latency for tasks and decrease throughput when doing filesystem IO through
+the pagecache. Direct reclaims are recorded using the allocstall counter
+in /proc/vmstat.
+
+The default value is 1 and the range of acceptible values are 1-16.
+Always start with lower values in the 2-6 range. Higher values should
+be justified with testing. If direct reclaims occur in spite of high
+values, the cost of direct reclaims (in latency) that occur can be
+higher due to increased lock contention.
 
 laptop_mode
 ===========

include/linux/mm.h

@@ -2309,6 +2309,7 @@ extern void set_dma_reserve(unsigned long new_dma_reserve);
 extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long,
 		enum memmap_context, struct vmem_altmap *);
 extern void setup_per_zone_wmarks(void);
+extern void update_kswapd_threads(void);
 extern int __meminit init_per_zone_wmark_min(void);
 extern void mem_init(void);
 extern void __init mmap_init(void);
@@ -2329,6 +2330,7 @@ extern void zone_pcp_update(struct zone *zone);
 extern void zone_pcp_reset(struct zone *zone);
 
 /* page_alloc.c */
+extern int kswapd_threads;
 extern int min_free_kbytes;
 extern int watermark_boost_factor;
 extern int watermark_scale_factor;
@@ -3017,5 +3019,12 @@ static inline int pages_identical(struct page *page1, struct page *page2)
 
 extern int want_old_faultaround_pte;
 
+#ifndef CONFIG_MULTIPLE_KSWAPD
+static inline void update_kswapd_threads_node(int nid) {}
+static inline int multi_kswapd_run(int nid) { return 0; }
+static inline void multi_kswapd_stop(int nid) {}
+static inline void multi_kswapd_cpu_online(pg_data_t *pgdat,
+					const struct cpumask *mask) {}
+#endif /* CONFIG_MULTIPLE_KSWAPD */
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */

include/linux/mmzone.h

@@ -39,6 +39,8 @@
  */
 #define PAGE_ALLOC_COSTLY_ORDER 3
 
+#define MAX_KSWAPD_THREADS 16
+
 enum migratetype {
 	MIGRATE_UNMOVABLE,
 	MIGRATE_MOVABLE,
@@ -743,8 +745,13 @@ typedef struct pglist_data {
 	int node_id;
 	wait_queue_head_t kswapd_wait;
 	wait_queue_head_t pfmemalloc_wait;
-	struct task_struct *kswapd;	/* Protected by
+	struct task_struct *kswapd;
-					   mem_hotplug_begin/end() */
+#ifdef CONFIG_MULTIPLE_KSWAPD
+	/*
+	 * Protected by mem_hotplug_begin/end()
+	 */
+	struct task_struct *mkswapd[MAX_KSWAPD_THREADS];
+#endif
 	int kswapd_order;
 	enum zone_type kswapd_classzone_idx;
 
@@ -957,6 +964,9 @@ static inline int is_highmem(struct zone *zone)
 
 /* These two functions are used to setup the per zone pages min values */
 struct ctl_table;
+int kswapd_threads_sysctl_handler(struct ctl_table *table, int write,
+					void __user *buffer, size_t *length,
+					loff_t *pos);
 int min_free_kbytes_sysctl_handler(struct ctl_table *, int,
 					void __user *, size_t *, loff_t *);
 int watermark_boost_factor_sysctl_handler(struct ctl_table *, int,

kernel/sysctl.c

@@ -143,6 +143,8 @@ static int six_hundred_forty_kb = 640 * 1024;
 static unsigned int __maybe_unused half_million = 500000;
 static unsigned int __maybe_unused one_hundred_million = 100000000;
 static unsigned int __maybe_unused one_million = 1000000;
+static int __maybe_unused max_kswapd_threads = MAX_KSWAPD_THREADS;
+
 #ifdef CONFIG_SCHED_WALT
 static int neg_three = -3;
 static int three = 3;
@@ -1825,6 +1827,17 @@ static struct ctl_table vm_table[] = {
 		.proc_handler	= watermark_boost_factor_sysctl_handler,
 		.extra1		= SYSCTL_ZERO,
 	},
+#ifdef CONFIG_MULTIPLE_KSWAPD
+	{
+		.procname	= "kswapd_threads",
+		.data		= &kswapd_threads,
+		.maxlen		= sizeof(kswapd_threads),
+		.mode		= 0644,
+		.proc_handler	= kswapd_threads_sysctl_handler,
+		.extra1		= SYSCTL_ONE,
+		.extra2		= &max_kswapd_threads,
+	},
+#endif
 	{
 		.procname	= "watermark_scale_factor",
 		.data		= &watermark_scale_factor,

mm/Kconfig

@@ -861,3 +861,16 @@ config PROCESS_RECLAIM
 	 (addr, addr + size-bytes) of the process.
 
 	 Any other value is ignored.
+
+config MULTIPLE_KSWAPD
+	bool "Spawn multiple kswapd threads"
+	depends on QGKI
+	default y
+	help
+	 kswapd_threads allows you to control the number of kswapd threads
+	 per node running on the system. The default value is 1 and the
+	 range of acceptible values are 1-16. The number of threads can
+	 be controlled by below command:
+	 (echo <num> > /proc/sys/vm/kswapd_threads)
+
+	 Values not in the range of 1..16 are ignored.

mm/page_alloc.c

@@ -8047,6 +8047,22 @@ int watermark_boost_factor_sysctl_handler(struct ctl_table *table, int write,
 	return 0;
 }
 
+#ifdef CONFIG_MULTIPLE_KSWAPD
+int kswapd_threads_sysctl_handler(struct ctl_table *table, int write,
+	void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int rc;
+
+	rc = proc_dointvec_minmax(table, write, buffer, length, ppos);
+	if (rc)
+		return rc;
+
+	if (write)
+		update_kswapd_threads();
+
+	return 0;
+}
+#endif
 int watermark_scale_factor_sysctl_handler(struct ctl_table *table, int write,
 	void __user *buffer, size_t *length, loff_t *ppos)
 {

mm/vmscan.c

@@ -139,6 +139,13 @@ struct scan_control {
 	struct vm_area_struct *target_vma;
 };
 
+/*
+ * Number of active kswapd threads
+ */
+#define DEF_KSWAPD_THREADS_PER_NODE 1
+int kswapd_threads = DEF_KSWAPD_THREADS_PER_NODE;
+int kswapd_threads_current = DEF_KSWAPD_THREADS_PER_NODE;
+
 #ifdef ARCH_HAS_PREFETCH
 #define prefetch_prev_lru_page(_page, _base, _field)			\
 	do {								\
@@ -4111,6 +4118,116 @@ unsigned long shrink_all_memory(unsigned long nr_to_reclaim)
 }
 #endif /* CONFIG_HIBERNATION */
 
+#ifdef CONFIG_MULTIPLE_KSWAPD
+static void update_kswapd_threads_node(int nid)
+{
+	pg_data_t *pgdat;
+	int drop, increase;
+	int last_idx, start_idx, hid;
+	int nr_threads = kswapd_threads_current;
+
+	pgdat = NODE_DATA(nid);
+	last_idx = nr_threads - 1;
+	if (kswapd_threads < nr_threads) {
+		drop = nr_threads - kswapd_threads;
+		for (hid = last_idx; hid > (last_idx - drop); hid--) {
+			if (pgdat->mkswapd[hid]) {
+				kthread_stop(pgdat->mkswapd[hid]);
+				pgdat->mkswapd[hid] = NULL;
+			}
+		}
+	} else {
+		increase = kswapd_threads - nr_threads;
+		start_idx = last_idx + 1;
+		for (hid = start_idx; hid < (start_idx + increase); hid++) {
+			pgdat->mkswapd[hid] = kthread_run(kswapd, pgdat,
+						"kswapd%d:%d", nid, hid);
+			if (IS_ERR(pgdat->mkswapd[hid])) {
+				pr_err("Failed to start kswapd%d on node %d\n",
+					hid, nid);
+				pgdat->mkswapd[hid] = NULL;
+				/*
+				 * We are out of resources. Do not start any
+				 * more threads.
+				 */
+				break;
+			}
+		}
+	}
+}
+
+void update_kswapd_threads(void)
+{
+	int nid;
+
+	if (kswapd_threads_current == kswapd_threads)
+		return;
+
+	/*
+	 * Hold the memory hotplug lock to avoid racing with memory
+	 * hotplug initiated updates
+	 */
+	mem_hotplug_begin();
+	for_each_node_state(nid, N_MEMORY)
+		update_kswapd_threads_node(nid);
+
+	pr_info("kswapd_thread count changed, old:%d new:%d\n",
+		kswapd_threads_current, kswapd_threads);
+	kswapd_threads_current = kswapd_threads;
+	mem_hotplug_done();
+}
+
+static int multi_kswapd_run(int nid)
+{
+	pg_data_t *pgdat = NODE_DATA(nid);
+	int hid, nr_threads = kswapd_threads;
+	int ret = 0;
+
+	pgdat->mkswapd[0] = pgdat->kswapd;
+	for (hid = 1; hid < nr_threads; ++hid) {
+		pgdat->mkswapd[hid] = kthread_run(kswapd, pgdat, "kswapd%d:%d",
+								nid, hid);
+		if (IS_ERR(pgdat->mkswapd[hid])) {
+			/* failure at boot is fatal */
+			WARN_ON(system_state < SYSTEM_RUNNING);
+			pr_err("Failed to start kswapd%d on node %d\n",
+				hid, nid);
+			ret = PTR_ERR(pgdat->mkswapd[hid]);
+			pgdat->mkswapd[hid] = NULL;
+		}
+	}
+	kswapd_threads_current = nr_threads;
+
+	return ret;
+}
+
+static void multi_kswapd_stop(int nid)
+{
+	int hid = 0;
+	int nr_threads = kswapd_threads_current;
+	struct task_struct *kswapd;
+
+	NODE_DATA(nid)->mkswapd[hid] = NULL;
+	for (hid = 1; hid < nr_threads; hid++) {
+		kswapd = NODE_DATA(nid)->mkswapd[hid];
+		if (kswapd) {
+			kthread_stop(kswapd);
+			NODE_DATA(nid)->mkswapd[hid] = NULL;
+		}
+	}
+}
+
+static void multi_kswapd_cpu_online(pg_data_t *pgdat,
+					const struct cpumask *mask)
+{
+	int hid;
+	int nr_threads = kswapd_threads_current;
+
+	for (hid = 1; hid < nr_threads; hid++)
+		set_cpus_allowed_ptr(pgdat->mkswapd[hid], mask);
+}
+#endif
+
 /* It's optimal to keep kswapds on the same CPUs as their memory, but
    not required for correctness.  So if the last cpu in a node goes
    away, we get changed to run anywhere: as the first one comes back,
@@ -4125,9 +4242,11 @@ static int kswapd_cpu_online(unsigned int cpu)
 
 		mask = cpumask_of_node(pgdat->node_id);
 
-		if (cpumask_any_and(cpu_online_mask, mask) < nr_cpu_ids)
+		if (cpumask_any_and(cpu_online_mask, mask) < nr_cpu_ids) {
 			/* One of our CPUs online: restore mask */
 			set_cpus_allowed_ptr(pgdat->kswapd, mask);
+			multi_kswapd_cpu_online(pgdat, mask);
+		}
 	}
 	return 0;
 }
@@ -4144,14 +4263,17 @@ int kswapd_run(int nid)
 	if (pgdat->kswapd)
 		return 0;
 
-	pgdat->kswapd = kthread_run(kswapd, pgdat, "kswapd%d", nid);
+	pgdat->kswapd = kthread_run(kswapd, pgdat, "kswapd%d:0", nid);
 	if (IS_ERR(pgdat->kswapd)) {
 		/* failure at boot is fatal */
 		BUG_ON(system_state < SYSTEM_RUNNING);
 		pr_err("Failed to start kswapd on node %d\n", nid);
 		ret = PTR_ERR(pgdat->kswapd);
 		pgdat->kswapd = NULL;
+		return ret;
 	}
+	ret = multi_kswapd_run(nid);
+
 	return ret;
 }
 
@@ -4167,6 +4289,8 @@ void kswapd_stop(int nid)
 		kthread_stop(kswapd);
 		NODE_DATA(nid)->kswapd = NULL;
 	}
+
+	multi_kswapd_stop(nid);
 }
 
 static int __init kswapd_init(void)