---

yaml
---
r: 254107
b: refs/heads/master
c: 0814257
h: refs/heads/master
i:
  254105: b964aa1
  254103: 0039840
v: v3

[refs]

@@ -1,2 +1,2 @@
 ---
-refs/heads/master: 9b679320a5fbf46454011e5c62e0b8991b0956d1
+refs/heads/master: 08142579b6ca35883c1ed066a2681de6f6917062

trunk/fs/inode.c

@@ -423,7 +423,14 @@ EXPORT_SYMBOL(remove_inode_hash);
 void end_writeback(struct inode *inode)
 {
 	might_sleep();
+	/*
+	 * We have to cycle tree_lock here because reclaim can be still in the
+	 * process of removing the last page (in __delete_from_page_cache())
+	 * and we must not free mapping under it.
+	 */
+	spin_lock_irq(&inode->i_data.tree_lock);
 	BUG_ON(inode->i_data.nrpages);
+	spin_unlock_irq(&inode->i_data.tree_lock);
 	BUG_ON(!list_empty(&inode->i_data.private_list));
 	BUG_ON(!(inode->i_state & I_FREEING));
 	BUG_ON(inode->i_state & I_CLEAR);

trunk/include/linux/fs.h

@@ -639,6 +639,7 @@ struct address_space {
 	struct prio_tree_root	i_mmap;		/* tree of private and shared mappings */
 	struct list_head	i_mmap_nonlinear;/*list VM_NONLINEAR mappings */
 	struct mutex		i_mmap_mutex;	/* protect tree, count, list */
+	/* Protected by tree_lock together with the radix tree */
 	unsigned long		nrpages;	/* number of total pages */
 	pgoff_t			writeback_index;/* writeback starts here */
 	const struct address_space_operations *a_ops;	/* methods */

trunk/mm/truncate.c

@@ -304,6 +304,11 @@ EXPORT_SYMBOL(truncate_inode_pages_range);
  * @lstart: offset from which to truncate
  *
  * Called under (and serialised by) inode->i_mutex.
+ *
+ * Note: When this function returns, there can be a page in the process of
+ * deletion (inside __delete_from_page_cache()) in the specified range.  Thus
+ * mapping->nrpages can be non-zero when this function returns even after
+ * truncation of the whole mapping.
  */
 void truncate_inode_pages(struct address_space *mapping, loff_t lstart)
 {