Changelog since v3
1. Added reclaim retry logic to avoid being OOM'ed due to pages from
   swap cache (coming in due to reclaim) don't overwhelm the container.

Changelog
1. Fix probable NULL pointer dereference based on review comments
   by YAMAMOTO Takashi

Add the meta_page to the per container LRU. The reclaim algorithm has been
modified to make the isolate_lru_pages() as a pluggable component. The
scan_control data structure now accepts the container on behalf of which
reclaims are carried out. try_to_free_pages() has been extended to become
container aware.

Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: <balbir@linux.vnet.ibm.com>
---

 include/linux/memcontrol.h  |   12 +++
 include/linux/res_counter.h |   23 +++++++
 include/linux/swap.h        |    3 
 mm/memcontrol.c             |  134 +++++++++++++++++++++++++++++++++++++++++++-
 mm/swap.c                   |    2 
 mm/vmscan.c                 |  126 +++++++++++++++++++++++++++++++++--------
 6 files changed, 274 insertions(+), 26 deletions(-)

diff -puN include/linux/memcontrol.h~mem-control-lru-and-reclaim include/linux/memcontrol.h
--- linux-2.6.23-rc1-mm1/include/linux/memcontrol.h~mem-control-lru-and-reclaim	2007-08-13 23:06:12.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/include/linux/memcontrol.h	2007-08-13 23:06:12.000000000 +0530
@@ -32,6 +32,13 @@ extern void page_assign_page_container(s
 extern struct page_container *page_get_page_container(struct page *page);
 extern int mem_container_charge(struct page *page, struct mm_struct *mm);
 extern void mem_container_uncharge(struct page_container *pc);
+extern void mem_container_move_lists(struct page_container *pc, bool active);
+extern unsigned long mem_container_isolate_pages(unsigned long nr_to_scan,
+					struct list_head *dst,
+					unsigned long *scanned, int order,
+					int mode, struct zone *z,
+					struct mem_container *mem_cont,
+					int active);
 
 static inline void mem_container_uncharge_page(struct page *page)
 {
@@ -71,6 +78,11 @@ static inline void mem_container_uncharg
 {
 }
 
+static inline void mem_container_move_lists(struct page_container *pc,
+						bool active)
+{
+}
+
 #endif /* CONFIG_CONTAINER_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff -puN include/linux/res_counter.h~mem-control-lru-and-reclaim include/linux/res_counter.h
--- linux-2.6.23-rc1-mm1/include/linux/res_counter.h~mem-control-lru-and-reclaim	2007-08-13 23:06:12.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/include/linux/res_counter.h	2007-08-13 23:06:12.000000000 +0530
@@ -99,4 +99,27 @@ int res_counter_charge(struct res_counte
 void res_counter_uncharge_locked(struct res_counter *counter, unsigned long val);
 void res_counter_uncharge(struct res_counter *counter, unsigned long val);
 
+static inline bool res_counter_limit_check_locked(struct res_counter *cnt)
+{
+	if (cnt->usage < cnt->limit)
+		return true;
+
+	return false;
+}
+
+/*
+ * Helper function to detect if the container is within it's limit or
+ * not. It's currently called from container_rss_prepare()
+ */
+static inline bool res_counter_check_under_limit(struct res_counter *cnt)
+{
+	bool ret;
+	unsigned long flags;
+
+	spin_lock_irqsave(&cnt->lock, flags);
+	ret = res_counter_limit_check_locked(cnt);
+	spin_unlock_irqrestore(&cnt->lock, flags);
+	return ret;
+}
+
 #endif
diff -puN include/linux/swap.h~mem-control-lru-and-reclaim include/linux/swap.h
--- linux-2.6.23-rc1-mm1/include/linux/swap.h~mem-control-lru-and-reclaim	2007-08-13 23:06:12.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/include/linux/swap.h	2007-08-13 23:06:12.000000000 +0530
@@ -6,6 +6,7 @@
 #include <linux/mmzone.h>
 #include <linux/list.h>
 #include <linux/sched.h>
+#include <linux/memcontrol.h>
 
 #include <asm/atomic.h>
 #include <asm/page.h>
@@ -191,6 +192,8 @@ extern void swap_setup(void);
 /* linux/mm/vmscan.c */
 extern unsigned long try_to_free_pages(struct zone **zones, int order,
 					gfp_t gfp_mask);
+extern unsigned long try_to_free_mem_container_pages(struct mem_container *mem);
+extern int __isolate_lru_page(struct page *page, int mode);
 extern unsigned long shrink_all_memory(unsigned long nr_pages);
 extern int vm_swappiness;
 extern int remove_mapping(struct address_space *mapping, struct page *page);
diff -puN mm/memcontrol.c~mem-control-lru-and-reclaim mm/memcontrol.c
--- linux-2.6.23-rc1-mm1/mm/memcontrol.c~mem-control-lru-and-reclaim	2007-08-13 23:06:12.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/mm/memcontrol.c	2007-08-13 23:06:12.000000000 +0530
@@ -24,8 +24,11 @@
 #include <linux/page-flags.h>
 #include <linux/bit_spinlock.h>
 #include <linux/rcupdate.h>
+#include <linux/swap.h>
+#include <linux/spinlock.h>
 
 struct container_subsys mem_container_subsys;
+static const int MEM_CONTAINER_RECLAIM_RETRIES = 5;
 
 /*
  * The memory controller data structure. The memory controller controls both
@@ -51,6 +54,10 @@ struct mem_container {
 	 */
 	struct list_head active_list;
 	struct list_head inactive_list;
+	/*
+	 * spin_lock to protect the per container LRU
+	 */
+	spinlock_t lru_lock;
 };
 
 /*
@@ -141,6 +148,94 @@ void __always_inline unlock_page_contain
 	bit_spin_unlock(PAGE_CONTAINER_LOCK_BIT, &page->page_container);
 }
 
+void __mem_container_move_lists(struct page_container *pc, bool active)
+{
+	if (active)
+		list_move(&pc->lru, &pc->mem_container->active_list);
+	else
+		list_move(&pc->lru, &pc->mem_container->inactive_list);
+}
+
+/*
+ * This routine assumes that the appropriate zone's lru lock is already held
+ */
+void mem_container_move_lists(struct page_container *pc, bool active)
+{
+	struct mem_container *mem;
+	if (!pc)
+		return;
+
+	mem = pc->mem_container;
+
+	spin_lock(&mem->lru_lock);
+	__mem_container_move_lists(pc, active);
+	spin_unlock(&mem->lru_lock);
+}
+
+unsigned long mem_container_isolate_pages(unsigned long nr_to_scan,
+					struct list_head *dst,
+					unsigned long *scanned, int order,
+					int mode, struct zone *z,
+					struct mem_container *mem_cont,
+					int active)
+{
+	unsigned long nr_taken = 0;
+	struct page *page;
+	unsigned long scan;
+	LIST_HEAD(pc_list);
+	struct list_head *src;
+	struct page_container *pc;
+
+	if (active)
+		src = &mem_cont->active_list;
+	else
+		src = &mem_cont->inactive_list;
+
+	spin_lock(&mem_cont->lru_lock);
+	for (scan = 0; scan < nr_to_scan && !list_empty(src); scan++) {
+		pc = list_entry(src->prev, struct page_container, lru);
+		page = pc->page;
+		VM_BUG_ON(!pc);
+
+		if (PageActive(page) && !active) {
+			__mem_container_move_lists(pc, true);
+			scan--;
+			continue;
+		}
+		if (!PageActive(page) && active) {
+			__mem_container_move_lists(pc, false);
+			scan--;
+			continue;
+		}
+
+		/*
+		 * Reclaim, per zone
+		 * TODO: make the active/inactive lists per zone
+		 */
+		if (page_zone(page) != z)
+			continue;
+
+		/*
+		 * Check if the meta page went away from under us
+		 */
+		if (!list_empty(&pc->lru))
+			list_move(&pc->lru, &pc_list);
+		else
+			continue;
+
+		if (__isolate_lru_page(page, mode) == 0) {
+			list_move(&page->lru, dst);
+			nr_taken++;
+		}
+	}
+
+	list_splice(&pc_list, src);
+	spin_unlock(&mem_cont->lru_lock);
+
+	*scanned = scan;
+	return nr_taken;
+}
+
 /*
  * Charge the memory controller for page usage.
  * Return
@@ -151,6 +246,8 @@ int mem_container_charge(struct page *pa
 {
 	struct mem_container *mem;
 	struct page_container *pc, *race_pc;
+	unsigned long flags;
+	unsigned long nr_retries = MEM_CONTAINER_RECLAIM_RETRIES;
 
 	/*
 	 * Should page_container's go to their own slab?
@@ -197,7 +294,32 @@ int mem_container_charge(struct page *pa
 	 * If we created the page_container, we should free it on exceeding
 	 * the container limit.
 	 */
-	if (res_counter_charge(&mem->res, 1)) {
+	while (res_counter_charge(&mem->res, 1)) {
+		if (try_to_free_mem_container_pages(mem))
+			continue;
+
+		/*
+ 		 * try_to_free_mem_container_pages() might not give us a full
+ 		 * picture of reclaim. Some pages are reclaimed and might be
+ 		 * moved to swap cache or just unmapped from the container.
+ 		 * Check the limit again to see if the reclaim reduced the
+ 		 * current usage of the container before giving up
+ 		 */
+		if (res_counter_check_under_limit(&mem->res))
+			continue;
+			/*
+			 * Since we control both RSS and cache, we end up with a
+			 * very interesting scenario where we end up reclaiming
+			 * memory (essentially RSS), since the memory is pushed
+			 * to swap cache, we eventually end up adding those
+			 * pages back to our list. Hence we give ourselves a
+			 * few chances before we fail
+			 */
+		else if (nr_retries--) {
+			congestion_wait(WRITE, HZ/10);
+			continue;
+		}
+
 		css_put(&mem->css);
 		goto free_pc;
 	}
@@ -220,6 +342,10 @@ int mem_container_charge(struct page *pa
 	pc->page = page;
 	page_assign_page_container(page, pc);
 
+	spin_lock_irqsave(&mem->lru_lock, flags);
+	list_add(&pc->lru, &mem->active_list);
+	spin_unlock_irqrestore(&mem->lru_lock, flags);
+
 done:
 	unlock_page_container(page);
 	return 0;
@@ -239,6 +365,7 @@ void mem_container_uncharge(struct page_
 {
 	struct mem_container *mem;
 	struct page *page;
+	unsigned long flags;
 
 	if (!pc)
 		return;
@@ -251,6 +378,10 @@ void mem_container_uncharge(struct page_
 		page_assign_page_container(page, NULL);
 		unlock_page_container(page);
 		res_counter_uncharge(&mem->res, 1);
+
+ 		spin_lock_irqsave(&mem->lru_lock, flags);
+ 		list_del_init(&pc->lru);
+ 		spin_unlock_irqrestore(&mem->lru_lock, flags);
 		kfree(pc);
 	}
 }
@@ -309,6 +440,7 @@ mem_container_create(struct container_su
 	res_counter_init(&mem->res);
 	INIT_LIST_HEAD(&mem->active_list);
 	INIT_LIST_HEAD(&