vmscan.c

 * Scan @zone's unevictable LRU lists to check for pages that have become
 * evictable.  Move those that have to @zone's inactive list where they
 * become candidates for reclaim, unless shrink_inactive_zone() decides
 * to reactivate them.  Pages that are still unevictable are rotated
 * back onto @zone's unevictable list.
 */
#define SCAN_UNEVICTABLE_BATCH_SIZE 16UL /* arbitrary lock hold batch size */
static void scan_zone_unevictable_pages(struct zone *zone)
{
	struct list_head *l_unevictable = &zone->lru[LRU_UNEVICTABLE].list;
	unsigned long scan;
	unsigned long nr_to_scan = zone_page_state(zone, NR_UNEVICTABLE);

	while (nr_to_scan > 0) {
		unsigned long batch_size = min(nr_to_scan,
						SCAN_UNEVICTABLE_BATCH_SIZE);

		spin_lock_irq(&zone->lru_lock);
		for (scan = 0;  scan < batch_size; scan++) {
			struct page *page = lru_to_page(l_unevictable);

			if (!trylock_page(page))
				continue;

			prefetchw_prev_lru_page(page, l_unevictable, flags);

			if (likely(PageLRU(page) && PageUnevictable(page)))
				check_move_unevictable_page(page, zone);

			unlock_page(page);
		}
		spin_unlock_irq(&zone->lru_lock);

		nr_to_scan -= batch_size;
	}
}


/**
 * scan_all_zones_unevictable_pages - scan all unevictable lists for evictable pages
 *
 * A really big hammer:  scan all zones' unevictable LRU lists to check for
 * pages that have become evictable.  Move those back to the zones'
 * inactive list where they become candidates for reclaim.
 * This occurs when, e.g., we have unswappable pages on the unevictable lists,
 * and we add swap to the system.  As such, it runs in the context of a task
 * that has possibly/probably made some previously unevictable pages
 * evictable.
 */
static void scan_all_zones_unevictable_pages(void)
{
	struct zone *zone;

	for_each_zone(zone) {
		scan_zone_unevictable_pages(zone);
	}
}

/*
 * scan_unevictable_pages [vm] sysctl handler.  On demand re-scan of
 * all nodes' unevictable lists for evictable pages
 */
unsigned long scan_unevictable_pages;

int scan_unevictable_handler(struct ctl_table *table, int write,
			   void __user *buffer,
			   size_t *length, loff_t *ppos)
{
	proc_doulongvec_minmax(table, write, buffer, length, ppos);

	if (write && *(unsigned long *)table->data)
		scan_all_zones_unevictable_pages();

	scan_unevictable_pages = 0;
	return 0;
}

#ifdef CONFIG_NUMA
/*
 * per node 'scan_unevictable_pages' attribute.  On demand re-scan of
 * a specified node's per zone unevictable lists for evictable pages.
 */

static ssize_t read_scan_unevictable_node(struct sys_device *dev,
					  struct sysdev_attribute *attr,
					  char *buf)
{
	return sprintf(buf, "0\n");	/* always zero; should fit... */
}

static ssize_t write_scan_unevictable_node(struct sys_device *dev,
					   struct sysdev_attribute *attr,
					const char *buf, size_t count)
{
	struct zone *node_zones = NODE_DATA(dev->id)->node_zones;
	struct zone *zone;
	unsigned long res;
	unsigned long req = strict_strtoul(buf, 10, &res);

	if (!req)
		return 1;	/* zero is no-op */

	for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) {
		if (!populated_zone(zone))
			continue;
		scan_zone_unevictable_pages(zone);
	}
	return 1;
}


static SYSDEV_ATTR(scan_unevictable_pages, S_IRUGO | S_IWUSR,
			read_scan_unevictable_node,
			write_scan_unevictable_node);

int scan_unevictable_register_node(struct node *node)
{
	return sysdev_create_file(&node->sysdev, &attr_scan_unevictable_pages);
}

void scan_unevictable_unregister_node(struct node *node)
{
	sysdev_remove_file(&node->sysdev, &attr_scan_unevictable_pages);
}
#endif