Changelog for v3

1. Fix a probable leak with meta_page's (pointed out by Paul Menage)
2. Introduce a wrapper around mem_container_uncharge for uncharging pages
   mem_container_uncharge_page()

Changelog

1. Improved error handling, uncharge on errors and check to see if we are
   leaking pages (review by YAMAMOTO Takashi)

Add the accounting hooks. The accounting is carried out for RSS and Page
Cache (unmapped) pages. There is now a common limit and accounting for both.
The RSS accounting is accounted at page_add_*_rmap() and page_remove_rmap()
time. Page cache is accounted at add_to_page_cache(),
__delete_from_page_cache(). Swap cache is also accounted for.

Each page's meta_page is protected with a bit in page flags, this makes
handling of race conditions involving simultaneous mappings of a page easier.
A reference count is kept in the meta_page to deal with cases where a page
might be unmapped from the RSS of all tasks, but still lives in the page
cache.

Credits go to Vaidyanathan Srinivasan for helping with reference counting work
of the meta page. Almost all of the page cache accounting code has help from
Vaidyanathan Srinivasan.

Signed-off-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: <balbir@linux.vnet.ibm.com>
---

 include/linux/memcontrol.h |   20 ++++++
 include/linux/page-flags.h |    3 
 mm/filemap.c               |   12 +++
 mm/memcontrol.c            |  139 ++++++++++++++++++++++++++++++++++++++++++++-
 mm/memory.c                |   44 ++++++++++++--
 mm/migrate.c               |    6 +
 mm/page_alloc.c            |    3 
 mm/rmap.c                  |   16 ++++-
 mm/swap_state.c            |   12 +++
 mm/swapfile.c              |   40 +++++++-----
 10 files changed, 270 insertions(+), 25 deletions(-)

diff -puN include/linux/memcontrol.h~mem-control-accounting include/linux/memcontrol.h
--- linux-2.6.23-rc1-mm1/include/linux/memcontrol.h~mem-control-accounting	2007-07-28 01:12:49.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/include/linux/memcontrol.h	2007-07-28 01:12:49.000000000 +0530
@@ -29,6 +29,13 @@ extern void mm_init_container(struct mm_
 extern void mm_free_container(struct mm_struct *mm);
 extern void page_assign_meta_page(struct page *page, struct meta_page *mp);
 extern struct meta_page *page_get_meta_page(struct page *page);
+extern int mem_container_charge(struct page *page, struct mm_struct *mm);
+extern void mem_container_uncharge(struct meta_page *mp);
+
+static inline void mem_container_uncharge_page(struct page *page)
+{
+	mem_container_uncharge(page_get_meta_page(page));
+}
 
 #else /* CONFIG_CONTAINER_MEM_CONT */
 static inline void mm_init_container(struct mm_struct *mm,
@@ -50,6 +57,19 @@ static inline struct meta_page *page_get
 	return NULL;
 }
 
+static inline int mem_container_charge(struct page *page, struct mm_struct *mm)
+{
+	return 0;
+}
+
+static inline void mem_container_uncharge(struct meta_page *mp)
+{
+}
+
+static inline void mem_container_uncharge_page(struct page *page)
+{
+}
+
 #endif /* CONFIG_CONTAINER_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff -puN include/linux/page-flags.h~mem-control-accounting include/linux/page-flags.h
--- linux-2.6.23-rc1-mm1/include/linux/page-flags.h~mem-control-accounting	2007-07-28 01:12:49.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/include/linux/page-flags.h	2007-07-28 01:12:49.000000000 +0530
@@ -98,6 +98,9 @@
 #define PG_checked		PG_owner_priv_1 /* Used by some filesystems */
 #define PG_pinned		PG_owner_priv_1	/* Xen pinned pagetable */
 
+#define PG_metapage		21	/* Used for checking if a meta_page */
+					/* is associated with a page	    */
+
 #if (BITS_PER_LONG > 32)
 /*
  * 64-bit-only flags build down from bit 31
diff -puN mm/filemap.c~mem-control-accounting mm/filemap.c
--- linux-2.6.23-rc1-mm1/mm/filemap.c~mem-control-accounting	2007-07-28 01:12:49.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/mm/filemap.c	2007-07-28 01:12:49.000000000 +0530
@@ -31,6 +31,7 @@
 #include <linux/syscalls.h>
 #include <linux/cpuset.h>
 #include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */
+#include <linux/memcontrol.h>
 #include "internal.h"
 
 /*
@@ -116,6 +117,7 @@ void __remove_from_page_cache(struct pag
 {
 	struct address_space *mapping = page->mapping;
 
+	mem_container_uncharge_page(page);
 	radix_tree_delete(&mapping->page_tree, page->index);
 	page->mapping = NULL;
 	mapping->nrpages--;
@@ -442,6 +444,11 @@ int add_to_page_cache(struct page *page,
 	int error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM);
 
 	if (error == 0) {
+
+		error = mem_container_charge(page, current->mm);
+		if (error)
+			goto out;
+
 		write_lock_irq(&mapping->tree_lock);
 		error = radix_tree_insert(&mapping->page_tree, offset, page);
 		if (!error) {
@@ -451,10 +458,13 @@ int add_to_page_cache(struct page *page,
 			page->index = offset;
 			mapping->nrpages++;
 			__inc_zone_page_state(page, NR_FILE_PAGES);
-		}
+		} else
+			mem_container_uncharge_page(page);
+
 		write_unlock_irq(&mapping->tree_lock);
 		radix_tree_preload_end();
 	}
+out:
 	return error;
 }
 EXPORT_SYMBOL(add_to_page_cache);
diff -puN mm/memcontrol.c~mem-control-accounting mm/memcontrol.c
--- linux-2.6.23-rc1-mm1/mm/memcontrol.c~mem-control-accounting	2007-07-28 01:12:49.000000000 +0530
+++ linux-2.6.23-rc1-mm1-balbir/mm/memcontrol.c	2007-07-28 01:12:49.000000000 +0530
@@ -21,6 +21,9 @@
 #include <linux/memcontrol.h>
 #include <linux/container.h>
 #include <linux/mm.h>
+#include <linux/page-flags.h>
+#include <linux/bit_spinlock.h>
+#include <linux/rcupdate.h>
 
 struct container_subsys mem_container_subsys;
 
@@ -31,7 +34,9 @@ struct container_subsys mem_container_su
  * to help the administrator determine what knobs to tune.
  *
  * TODO: Add a water mark for the memory controller. Reclaim will begin when
- * we hit the water mark.
+ * we hit the water mark. May be even add a low water mark, such that
+ * no reclaim occurs from a container at it's low water mark, this is
+ * a feature that will be implemented much later in the future.
  */
 struct mem_container {
 	struct container_subsys_state css;
@@ -56,6 +61,8 @@ struct meta_page {
 	struct list_head lru;		/* per container LRU list */
 	struct page *page;
 	struct mem_container *mem_container;
+	atomic_t ref_cnt;		/* Helpful when pages move b/w  */
+					/* mapped and cached states     */
 };
 
 
@@ -98,6 +105,134 @@ struct meta_page *page_get_meta_page(str
 	return page->meta_page;
 }
 
+void __always_inline lock_meta_page(struct page *page)
+{
+	bit_spin_lock(PG_metapage, &page->flags);
+	VM_BUG_ON(!test_bit(PG_metapage, &page->flags));
+}
+
+void __always_inline unlock_meta_page(struct page *page)
+{
+	bit_spin_unlock(PG_metapage, &page->flags);
+}
+
+/*
+ * Charge the memory controller for page usage.
+ * Return
+ * 0 if the charge was successful
+ * < 0 if the container is over its limit
+ */
+int mem_container_charge(struct page *page, struct mm_struct *mm)
+{
+	struct mem_container *mem;
+	struct meta_page *mp, *race_mp;
+
+	/*
+	 * Should meta_page's go to their own slab?
+	 * One could optimize the performance of the charging routine
+	 * by saving a bit in the page_flags and using it as a lock
+	 * to see if the container page already has a meta_page associated
+	 * with it
+	 */
+	lock_meta_page(page);
+	mp = page_get_meta_page(page);
+	/*
+	 * The meta_page exists and the page has already been accounted
+	 */
+	if (mp) {
+		atomic_inc(&mp->ref_cnt);
+		goto done;
+	}
+
+	unlock_meta_page(page);
+
+	mp = kzalloc(sizeof(struct meta_page), GFP_KERNEL);
+	if (mp == NULL)
+		goto err;
+
+	rcu_read_lock();
+	/*
+	 * We always charge the container the mm_struct belongs to
+	 * the mm_struct's mem_container changes on task migration if the
+	 * thread group leader migrates. It's possible that mm is not
+	 * set, if so charge the init_mm (happens for pagecache usage).
+	 */
+	if (!mm)
+		mm = &init_mm;
+
+	mem = rcu_dereference(mm->mem_container);
+	/*
+	 * For every charge from the container, increment reference
+	 * count
+	 */
+	css_get(&mem->css);
+	rcu_read_unlock();
+
+	/*
+	 * If we created the meta_page, we should free it on exceeding
+	 * the container limit.
+	 */
+	if (res_counter_charge(&mem->res, 1)) {
+		css_put(&mem->css);
+		goto free_mp;
+	}
+
+	lock_meta_page(page);
+	/*
+	 * Check if somebody else beat us to allocating the meta_page
+	 */
+	race_mp = page_get_meta_page(page);
+	if (race_mp) {
+		kfree(mp);
+		mp = race_mp;
+		atomic_inc(&mp->ref_cnt);
+		res_counter_uncharge(&mem->res, 1);
+		goto done;
+	}
+
+	atomic_set(&mp->ref_cnt, 1);
+	mp->mem_container = mem;
+	mp->page = page;
+	page_assign_meta_page(page, mp);
+
+done:
+	unlock_meta_page(page);
+	return 0;
+free_mp:
+	kfree(mp);
+	return -ENOMEM;
+err:
+	unlock_meta_page(page);
+	return -ENOMEM;
+}
+
+/*
+ * Uncharging is always a welcome operation, we never complain, simply
+ * uncharge.
+ */
+void mem_container_uncharge(struct meta_page *mp)
+{
+	struct mem_container *mem;
+	struct page *page;
+
+	/*
+	 * This can happen for PAGE_ZERO
+	 */
+	if (!mp)
+		return;
+
+	if (atomic_dec_and_test(&mp->ref_cnt)) {
+		page = mp->page;
+		lock_meta_page(page);
+		mem = mp->mem_container;
+		css_put(&mem->css);
+		page_assign_meta_page(page, NULL);
+		unlock_meta_page(page);
+		res_counter_uncharge(&mem->res, 1);
+		kfree(mp);
+	}
+}
+
 static ssize_t mem_container_read(struct container *cont, struct cftype *cft,
 			struct file *file, char __user *userbuf, size_t nbytes,
 			loff_t *ppos)
@@ -150,6 +285,8 @@ mem_container_create(struct container_su
 		return NULL;
 
 	res_counter_init(&mem->res);
+	INIT_LIST_HEAD(&mem->active_list);
+	INIT_LIST_HEAD(&mem->inactive_list);
 	return &mem->css;
 }
 
diff -puN mm/memor