memcg: bad page if page_cgroup when free

Replace free_hot_cold_page's VM_BUG_ON(page_get_page_cgroup(page)) by a "Bad
page state" and clear: most users don't have CONFIG_DEBUG_VM on, and if it
were set here, it'd likely cause corruption when the page is reused.

Don't use page_assign_page_cgroup to clear it: that should be private to
memcontrol.c, and always called with the lock taken; and memmap_init_zone
doesn't need it either - like page->mapping and other pointers throughout the
kernel, Linux assumes pointers in zeroed structures are NULL pointers.

Instead use page_reset_bad_cgroup, added to memcontrol.h for this only.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: David Rientjes <rientjes@google.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hirokazu Takahashi <taka@valinux.co.jp>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

include/linux/memcontrol.h

@@ -29,8 +29,9 @@ struct mm_struct;
 
 extern void mm_init_cgroup(struct mm_struct *mm, struct task_struct *p);
 extern void mm_free_cgroup(struct mm_struct *mm);
-extern void page_assign_page_cgroup(struct page *page,
-					struct page_cgroup *pc);
+
+#define page_reset_bad_cgroup(page)	((page)->page_cgroup = 0)
+
 extern struct page_cgroup *page_get_page_cgroup(struct page *page);
 extern int mem_cgroup_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask);
@@ -82,8 +83,7 @@ static inline void mm_free_cgroup(struct mm_struct *mm)
 {
 }
 
-static inline void page_assign_page_cgroup(struct page *page,
-						struct page_cgroup *pc)
+static inline void page_reset_bad_cgroup(struct page *page)
 {
 }
 

mm/memcontrol.c

@@ -140,11 +140,17 @@ struct mem_cgroup {
 
 /*
  * We use the lower bit of the page->page_cgroup pointer as a bit spin
- * lock. We need to ensure that page->page_cgroup is atleast two
- * byte aligned (based on comments from Nick Piggin)
+ * lock.  We need to ensure that page->page_cgroup is at least two
+ * byte aligned (based on comments from Nick Piggin).  But since
+ * bit_spin_lock doesn't actually set that lock bit in a non-debug
+ * uniprocessor kernel, we should avoid setting it here too.
  */
 #define PAGE_CGROUP_LOCK_BIT 	0x0
-#define PAGE_CGROUP_LOCK 		(1 << PAGE_CGROUP_LOCK_BIT)
+#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
+#define PAGE_CGROUP_LOCK 	(1 << PAGE_CGROUP_LOCK_BIT)
+#else
+#define PAGE_CGROUP_LOCK	0x0
+#endif
 
 /*
  * A page_cgroup page is associated with every page descriptor. The
@@ -271,19 +277,10 @@ static inline int page_cgroup_locked(struct page *page)
 					&page->page_cgroup);
 }
 
-void page_assign_page_cgroup(struct page *page, struct page_cgroup *pc)
+static void page_assign_page_cgroup(struct page *page, struct page_cgroup *pc)
 {
-	int locked;
-
-	/*
-	 * While resetting the page_cgroup we might not hold the
-	 * page_cgroup lock. free_hot_cold_page() is an example
-	 * of such a scenario
-	 */
-	if (pc)
-		VM_BUG_ON(!page_cgroup_locked(page));
-	locked = (page->page_cgroup & PAGE_CGROUP_LOCK);
-	page->page_cgroup = ((unsigned long)pc | locked);
+	VM_BUG_ON(!page_cgroup_locked(page));
+	page->page_cgroup = ((unsigned long)pc | PAGE_CGROUP_LOCK);
 }
 
 struct page_cgroup *page_get_page_cgroup(struct page *page)

mm/page_alloc.c

@@ -222,13 +222,19 @@ static inline int bad_range(struct zone *zone, struct page *page)
 
 static void bad_page(struct page *page)
 {
-	printk(KERN_EMERG "Bad page state in process '%s'\n"
-		KERN_EMERG "page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d\n"
-		KERN_EMERG "Trying to fix it up, but a reboot is needed\n"
-		KERN_EMERG "Backtrace:\n",
+	void *pc = page_get_page_cgroup(page);
+
+	printk(KERN_EMERG "Bad page state in process '%s'\n" KERN_EMERG
+		"page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d\n",
 		current->comm, page, (int)(2*sizeof(unsigned long)),
 		(unsigned long)page->flags, page->mapping,
 		page_mapcount(page), page_count(page));
+	if (pc) {
+		printk(KERN_EMERG "cgroup:%p\n", pc);
+		page_reset_bad_cgroup(page);
+	}
+	printk(KERN_EMERG "Trying to fix it up, but a reboot is needed\n"
+		KERN_EMERG "Backtrace:\n");
 	dump_stack();
 	page->flags &= ~(1 << PG_lru	|
 			1 << PG_private |
@@ -454,6 +460,7 @@ static inline int free_pages_check(struct page *page)
 {
 	if (unlikely(page_mapcount(page) |
 		(page->mapping != NULL)  |
+		(page_get_page_cgroup(page) != NULL) |
 		(page_count(page) != 0)  |
 		(page->flags & (
 			1 << PG_lru	|
@@ -603,6 +610,7 @@ static int prep_new_page(struct page *page, int order, gfp_t gfp_flags)
 {
 	if (unlikely(page_mapcount(page) |
 		(page->mapping != NULL)  |
+		(page_get_page_cgroup(page) != NULL) |
 		(page_count(page) != 0)  |
 		(page->flags & (
 			1 << PG_lru	|
@@ -989,7 +997,6 @@ static void free_hot_cold_page(struct page *page, int cold)
 
 	if (!PageHighMem(page))
 		debug_check_no_locks_freed(page_address(page), PAGE_SIZE);
-	VM_BUG_ON(page_get_page_cgroup(page));
 	arch_free_page(page, 0);
 	kernel_map_pages(page, 1, 0);
 
@@ -2528,7 +2535,6 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone,
 		set_page_links(page, zone, nid, pfn);
 		init_page_count(page);
 		reset_page_mapcount(page);
-		page_assign_page_cgroup(page, NULL);
 		SetPageReserved(page);
 
 		/*