[PATCH 4/6] locks: break delegations on rename

["J. Bruce Fields" <bfields@redhat.com>]

From: "J. Bruce Fields" <bfields@redhat.com>

To rely on the i_mutex for exclusion between setlease and rename, we
need rename to take the i_mutex on the source as well as on any possible
target.

Also fix up lockdep and the Documentation/filesystems/directory-locking
documentation.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 Documentation/filesystems/directory-locking |   11 ++++++-----
 fs/namei.c                                  |   12 ++++++++++++
 include/linux/fs.h                          |    9 ++++++---
 3 files changed, 24 insertions(+), 8 deletions(-)

diff --git a/Documentation/filesystems/directory-locking b/Documentation/filesystems/directory-locking
index ff7b611..9edbcd2 100644
--- a/Documentation/filesystems/directory-locking
+++ b/Documentation/filesystems/directory-locking
@@ -12,8 +12,8 @@ kinds of locks - per-inode (->i_mutex) and per-filesystem
 locks victim and calls the method.
 
 4) rename() that is _not_ cross-directory.  Locking rules: caller locks
-the parent, finds source and target, if target already exists - locks it
-and then calls the method.
+the parent, finds source and target, locks source, also locks target if
+it already exists, and then calls the method.
 
 5) link creation.  Locking rules:
 	* lock parent
@@ -30,6 +30,7 @@ rules:
 		fail with -ENOTEMPTY
 	* if new parent is equal to or is a descendent of source
 		fail with -ELOOP
+	* lock source if it is not a directory.
 	* if target exists - lock it.
 	* call the method.
 
@@ -56,9 +57,9 @@ objects - A < B iff A is an ancestor of B.
     renames will be blocked on filesystem lock and we don't start changing
     the order until we had acquired all locks).
 
-(3) any operation holds at most one lock on non-directory object and
-    that lock is acquired after all other locks.  (Proof: see descriptions
-    of operations).
+(3) locks on non-directory objects are acquired only after taking locks
+    on their parents (which remain their parents until all locks are
+    acquired, by (1) and (2)).  (Proof: see descriptions of operations).
 
 	Now consider the minimal deadlock.  Each process is blocked on
 attempt to acquire some lock and already holds at least one lock.  Let's
diff --git a/fs/namei.c b/fs/namei.c
index 4516958..1e0c383 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -3076,6 +3076,7 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
 			    struct inode *new_dir, struct dentry *new_dentry)
 {
 	struct inode *target = new_dentry->d_inode;
+	struct inode *source = old_dentry->d_inode;
 	int error;
 
 	error = security_inode_rename(old_dir, old_dentry, new_dir, new_dentry);
@@ -3083,6 +3084,7 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
 		return error;
 
 	dget(new_dentry);
+	mutex_lock_nested(&source->i_mutex, I_MUTEX_RENAME_SOURCE);
 	if (target)
 		mutex_lock(&target->i_mutex);
 
@@ -3090,6 +3092,14 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
 	if (d_mountpoint(old_dentry)||d_mountpoint(new_dentry))
 		goto out;
 
+	error = break_lease(source, BREAK_ONLY_DELEGS|BREAK_R_AND_W_LEASES);
+	if (error)
+		goto out;
+	if (target) {
+		error = break_lease(target, BREAK_ONLY_DELEGS|BREAK_R_AND_W_LEASES);
+		if (error)
+			goto out;
+	}
 	error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry);
 	if (error)
 		goto out;
@@ -3101,6 +3111,8 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
 out:
 	if (target)
 		mutex_unlock(&target->i_mutex);
+out_unlock_source:
+	mutex_unlock(&source->i_mutex);
 	dput(new_dentry);
 	return error;
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 89c8481..f6a09b4 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -855,10 +855,12 @@ static inline int inode_unhashed(struct inode *inode)
  * 0: the object of the current VFS operation
  * 1: parent
  * 2: child/target
- * 3: quota file
+ * 3: xattr
+ * 4: quota file
+ * 5: the file being renamed (used only in rename of a non-directory)
  *
  * The locking order between these classes is
- * parent -> child -> normal -> xattr -> quota
+ * parent -> child -> rename_source -> normal -> xattr -> quota
  */
 enum inode_i_mutex_lock_class
 {
@@ -866,7 +868,8 @@ enum inode_i_mutex_lock_class
 	I_MUTEX_PARENT,
 	I_MUTEX_CHILD,
 	I_MUTEX_XATTR,
-	I_MUTEX_QUOTA
+	I_MUTEX_QUOTA,
+	I_MUTEX_RENAME_SOURCE
 };
 
 /*
-- 
1.7.5.4