new primitive: discard_new_inode()

We don't want open-by-handle picking half-set-up in-core
struct inode from e.g. mkdir() having failed halfway through.
In other words, we don't want such inodes returned by iget_locked()
on their way to extinction. However, we can't just have them
unhashed - otherwise open-by-handle immediately *after* that would've
ended up creating a new in-core inode over the on-disk one that
is in process of being freed right under us.

Solution: new flag (I_CREATING) set by insert_inode_locked() and
removed by unlock_new_inode() and a new primitive (discard_new_inode())
to be used by such halfway-through-setup failure exits instead of
unlock_new_inode() / iput() combinations. That primitive unlocks new
inode, but leaves I_CREATING in place.

iget_locked() treats finding an I_CREATING inode as failure
(-ESTALE, once we sort out the error propagation).
insert_inode_locked() treats the same as instant -EBUSY.
ilookup() treats those as icache miss.

[Fix by Dan Carpenter <[e-mail]> folded in]

Signed-off-by: Al Viro <[e-mail]>

fs/dcache.c

···
 	spin_lock(&inode->i_lock);
 	__d_instantiate(entry, inode);
 	WARN_ON(!(inode->i_state & I_NEW));
-	inode->i_state &= ~I_NEW;
+	inode->i_state &= ~I_NEW & ~I_CREATING;
 	smp_mb();
 	wake_up_bit(&inode->i_state, __I_NEW);
 	spin_unlock(&inode->i_lock);

fs/inode.c

···
 			__wait_on_freeing_inode(inode);
 			goto repeat;
 		}
+		if (unlikely(inode->i_state & I_CREATING)) {
+			spin_unlock(&inode->i_lock);
+			return ERR_PTR(-ESTALE);
+		}
 		__iget(inode);
 		spin_unlock(&inode->i_lock);
 		return inode;
···
 		if (inode->i_state & (I_FREEING|I_WILL_FREE)) {
 			__wait_on_freeing_inode(inode);
 			goto repeat;
+		}
+		if (unlikely(inode->i_state & I_CREATING)) {
+			spin_unlock(&inode->i_lock);
+			return ERR_PTR(-ESTALE);
 		}
 		__iget(inode);
 		spin_unlock(&inode->i_lock);
···
 	lockdep_annotate_inode_mutex_key(inode);
 	spin_lock(&inode->i_lock);
 	WARN_ON(!(inode->i_state & I_NEW));
-	inode->i_state &= ~I_NEW;
+	inode->i_state &= ~I_NEW & ~I_CREATING;
 	smp_mb();
 	wake_up_bit(&inode->i_state, __I_NEW);
 	spin_unlock(&inode->i_lock);
 }
 EXPORT_SYMBOL(unlock_new_inode);
+
+void discard_new_inode(struct inode *inode)
+{
+	lockdep_annotate_inode_mutex_key(inode);
+	spin_lock(&inode->i_lock);
+	WARN_ON(!(inode->i_state & I_NEW));
+	inode->i_state &= ~I_NEW;
+	smp_mb();
+	wake_up_bit(&inode->i_state, __I_NEW);
+	spin_unlock(&inode->i_lock);
+	iput(inode);
+}
+EXPORT_SYMBOL(discard_new_inode);
 
 /**
  * lock_two_nondirectories - take two i_mutexes on non-directory objects
···
 		 * Use the old inode instead of the preallocated one.
 		 */
 		spin_unlock(&inode_hash_lock);
+		if (IS_ERR(old))
+			return NULL;
 		wait_on_inode(old);
 		if (unlikely(inode_unhashed(old))) {
 			iput(old);
···
 	inode = find_inode_fast(sb, head, ino);
 	spin_unlock(&inode_hash_lock);
 	if (inode) {
+		if (IS_ERR(inode))
+			return NULL;
 		wait_on_inode(inode);
 		if (unlikely(inode_unhashed(inode))) {
 			iput(inode);
···
 		 */
 		spin_unlock(&inode_hash_lock);
 		destroy_inode(inode);
+		if (IS_ERR(old))
+			return NULL;
 		inode = old;
 		wait_on_inode(inode);
 		if (unlikely(inode_unhashed(inode))) {
···
 	inode = find_inode(sb, head, test, data);
 	spin_unlock(&inode_hash_lock);
 
-	return inode;
+	return IS_ERR(inode) ? NULL : inode;
 }
 EXPORT_SYMBOL(ilookup5_nowait);
 
···
 	spin_unlock(&inode_hash_lock);
 
 	if (inode) {
+		if (IS_ERR(inode))
+			return NULL;
 		wait_on_inode(inode);
 		if (unlikely(inode_unhashed(inode))) {
 			iput(inode);
···
 		}
 		if (likely(!old)) {
 			spin_lock(&inode->i_lock);
-			inode->i_state |= I_NEW;
+			inode->i_state |= I_NEW | I_CREATING;
 			hlist_add_head(&inode->i_hash, head);
 			spin_unlock(&inode->i_lock);
 			spin_unlock(&inode_hash_lock);
 			return 0;
+		}
+		if (unlikely(old->i_state & I_CREATING)) {
+			spin_unlock(&old->i_lock);
+			spin_unlock(&inode_hash_lock);
+			return -EBUSY;
 		}
 		__iget(old);
 		spin_unlock(&old->i_lock);
···
 int insert_inode_locked4(struct inode *inode, unsigned long hashval,
 		int (*test)(struct inode *, void *), void *data)
 {
-	struct inode *old = inode_insert5(inode, hashval, test, NULL, data);
+	struct inode *old;
+
+	inode->i_state |= I_CREATING;
+	old = inode_insert5(inode, hashval, test, NULL, data);
 
 	if (old != inode) {
 		iput(old);

include/linux/fs.h

···
  * I_OVL_INUSE		Used by overlayfs to get exclusive ownership on upper
  *			and work dirs among overlayfs mounts.
  *
+ * I_CREATING		New object's inode in the middle of setting up.
+ *
  * Q: What is the difference between I_WILL_FREE and I_FREEING?
  */
 #define I_DIRTY_SYNC		(1 << 0)
···
 #define __I_DIRTY_TIME_EXPIRED	12
 #define I_DIRTY_TIME_EXPIRED	(1 << __I_DIRTY_TIME_EXPIRED)
 #define I_WB_SWITCH		(1 << 13)
-#define I_OVL_INUSE			(1 << 14)
+#define I_OVL_INUSE		(1 << 14)
+#define I_CREATING		(1 << 15)
 
 #define I_DIRTY_INODE (I_DIRTY_SYNC | I_DIRTY_DATASYNC)
 #define I_DIRTY (I_DIRTY_INODE | I_DIRTY_PAGES)
···
 static inline void lockdep_annotate_inode_mutex_key(struct inode *inode) { };
 #endif
 extern void unlock_new_inode(struct inode *);
+extern void discard_new_inode(struct inode *);
 extern unsigned int get_next_ino(void);
 extern void evict_inodes(struct super_block *sb);