Saner locking around deactivate_super()

Make sure that s_umount is acquired *before* we drop the final
active reference; we still have the fast path (atomic_dec_unless)
and we have gotten rid of the window between the moment when
s_active hits zero and s_umount is acquired.  Which simplifies
the living hell out of grab_super() and inotify pin_to_kill()
stuff.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

fs/notify/inotify/inotify.c

@@ -511,34 +511,8 @@ EXPORT_SYMBOL_GPL(inotify_init_watch);
  * done.  Cleanup is just deactivate_super().  However, that leaves a messy
  * case - what if we *are* racing with umount() and active references to
  * superblock can't be acquired anymore?  We can bump ->s_count, grab
- * ->s_umount, which will almost certainly wait until the superblock is shut
- * down and the watch in question is pining for fjords.  That's fine, but
- * there is a problem - we might have hit the window between ->s_active
- * getting to 0 (i.e. the moment when superblock is past the point of no return
- * and is heading for shutdown) and the moment when deactivate_super() acquires
- * ->s_umount.  We could just do drop_super() yield() and retry, but that's
- * rather antisocial and this stuff is luser-triggerable.  OTOH, having grabbed
- * ->s_umount and having found that we'd got there first (i.e. that ->s_root is
- * non-NULL) we know that we won't race with inotify_umount_inodes().  So we
- * could grab a reference to watch and do the rest as above, just with
- * drop_super() instead of deactivate_super(), right?  Wrong.  We had to drop
- * ih->mutex before we could grab ->s_umount.  So the watch could've been gone
- * already.
- *
- * That still can be dealt with - we need to save watch->wd, do idr_find()
- * and compare its result with our pointer.  If they match, we either have
- * the damn thing still alive or we'd lost not one but two races at once,
- * the watch had been killed and a new one got created with the same ->wd
- * at the same address.  That couldn't have happened in inotify_destroy(),
- * but inotify_rm_wd() could run into that.  Still, "new one got created"
- * is not a problem - we have every right to kill it or leave it alone,
- * whatever's more convenient.
- *
- * So we can use idr_find(...) == watch && watch->inode->i_sb == sb as
- * "grab it and kill it" check.  If it's been our original watch, we are
- * fine, if it's a newcomer - nevermind, just pretend that we'd won the
- * race and kill the fscker anyway; we are safe since we know that its
- * superblock won't be going away.
+ * ->s_umount, which will wait until the superblock is shut down and the
+ * watch in question is pining for fjords.
  *
  * And yes, this is far beyond mere "not very pretty"; so's the entire
  * concept of inotify to start with.
@@ -552,14 +526,10 @@ EXPORT_SYMBOL_GPL(inotify_init_watch);
  * Called with ih->mutex held, drops it.  Possible return values:
  * 0 - nothing to do, it has died
  * 1 - remove it, drop the reference and deactivate_super()
- * 2 - remove it, drop the reference and drop_super(); we tried hard to avoid
- * that variant, since it involved a lot of PITA, but that's the best that
- * could've been done.
  */
 static int pin_to_kill(struct inotify_handle *ih, struct inotify_watch *watch)
 {
 	struct super_block *sb = watch->inode->i_sb;
-	s32 wd = watch->wd;
 
 	if (atomic_inc_not_zero(&sb->s_active)) {
 		get_inotify_watch(watch);
@@ -571,36 +541,16 @@ static int pin_to_kill(struct inotify_handle *ih, struct inotify_watch *watch)
 	spin_unlock(&sb_lock);
 	mutex_unlock(&ih->mutex); /* can't grab ->s_umount under it */
 	down_read(&sb->s_umount);
-	if (likely(!sb->s_root)) {
-		/* fs is already shut down; the watch is dead */
-		drop_super(sb);
-		return 0;
-	}
-	/* raced with the final deactivate_super() */
-	mutex_lock(&ih->mutex);
-	if (idr_find(&ih->idr, wd) != watch || watch->inode->i_sb != sb) {
-		/* the watch is dead */
-		mutex_unlock(&ih->mutex);
-		drop_super(sb);
-		return 0;
-	}
-	/* still alive or freed and reused with the same sb and wd; kill */
-	get_inotify_watch(watch);
-	mutex_unlock(&ih->mutex);
-	return 2;
+	/* fs is already shut down; the watch is dead */
+	drop_super(sb);
+	return 0;
 }
 
-static void unpin_and_kill(struct inotify_watch *watch, int how)
+static void unpin_and_kill(struct inotify_watch *watch)
 {
 	struct super_block *sb = watch->inode->i_sb;
 	put_inotify_watch(watch);
-	switch (how) {
-	case 1:
-		deactivate_super(sb);
-		break;
-	case 2:
-		drop_super(sb);
-	}
+	deactivate_super(sb);
 }
 
 /**
@@ -622,7 +572,6 @@ void inotify_destroy(struct inotify_handle *ih)
 		struct list_head *watches;
 		struct super_block *sb;
 		struct inode *inode;
-		int how;
 
 		mutex_lock(&ih->mutex);
 		watches = &ih->watches;
@@ -632,8 +581,7 @@ void inotify_destroy(struct inotify_handle *ih)
 		}
 		watch = list_first_entry(watches, struct inotify_watch, h_list);
 		sb = watch->inode->i_sb;
-		how = pin_to_kill(ih, watch);
-		if (!how)
+		if (!pin_to_kill(ih, watch))
 			continue;
 
 		inode = watch->inode;
@@ -648,7 +596,7 @@ void inotify_destroy(struct inotify_handle *ih)
 
 		mutex_unlock(&ih->mutex);
 		mutex_unlock(&inode->inotify_mutex);
-		unpin_and_kill(watch, how);
+		unpin_and_kill(watch);
 	}
 
 	/* free this handle: the put matching the get in inotify_init() */
@@ -851,7 +799,6 @@ int inotify_rm_wd(struct inotify_handle *ih, u32 wd)
 	struct inotify_watch *watch;
 	struct super_block *sb;
 	struct inode *inode;
-	int how;
 
 	mutex_lock(&ih->mutex);
 	watch = idr_find(&ih->idr, wd);
@@ -860,8 +807,7 @@ int inotify_rm_wd(struct inotify_handle *ih, u32 wd)
 		return -EINVAL;
 	}
 	sb = watch->inode->i_sb;
-	how = pin_to_kill(ih, watch);
-	if (!how)
+	if (!pin_to_kill(ih, watch))
 		return 0;
 
 	inode = watch->inode;
@@ -875,7 +821,7 @@ int inotify_rm_wd(struct inotify_handle *ih, u32 wd)
 
 	mutex_unlock(&ih->mutex);
 	mutex_unlock(&inode->inotify_mutex);
-	unpin_and_kill(watch, how);
+	unpin_and_kill(watch);
 
 	return 0;
 }

fs/super.c

@@ -178,53 +178,48 @@ void put_super(struct super_block *sb)
 
 
 /**
- *	deactivate_super	-	drop an active reference to superblock
+ *	deactivate_locked_super	-	drop an active reference to superblock
  *	@s: superblock to deactivate
  *
- *	Drops an active reference to superblock, acquiring a temprory one if
- *	there is no active references left.  In that case we lock superblock,
+ *	Drops an active reference to superblock, converting it into a temprory
+ *	one if there is no other active references left.  In that case we
  *	tell fs driver to shut it down and drop the temporary reference we
  *	had just acquired.
+ *
+ *	Caller holds exclusive lock on superblock; that lock is released.
  */
-void deactivate_super(struct super_block *s)
+void deactivate_locked_super(struct super_block *s)
 {
 	struct file_system_type *fs = s->s_type;
 	if (atomic_dec_and_test(&s->s_active)) {
 		vfs_dq_off(s, 0);
-		down_write(&s->s_umount);
 		fs->kill_sb(s);
 		put_filesystem(fs);
 		put_super(s);
+	} else {
+		up_write(&s->s_umount);
 	}
 }
 
-EXPORT_SYMBOL(deactivate_super);
+EXPORT_SYMBOL(deactivate_locked_super);
 
 /**
- *	deactivate_locked_super	-	drop an active reference to superblock
+ *	deactivate_super	-	drop an active reference to superblock
  *	@s: superblock to deactivate
  *
- *	Equivalent of up_write(&s->s_umount); deactivate_super(s);, except that
- *	it does not unlock it until it's all over.  As the result, it's safe to
- *	use to dispose of new superblock on ->get_sb() failure exits - nobody
- *	will see the sucker until it's all over.  Equivalent using up_write +
- *	deactivate_super is safe for that purpose only if superblock is either
- *	safe to use or has NULL ->s_root when we unlock.
+ *	Variant of deactivate_locked_super(), except that superblock is *not*
+ *	locked by caller.  If we are going to drop the final active reference,
+ *	lock will be acquired prior to that.
  */
-void deactivate_locked_super(struct super_block *s)
+void deactivate_super(struct super_block *s)
 {
-	struct file_system_type *fs = s->s_type;
-	if (atomic_dec_and_test(&s->s_active)) {
-		vfs_dq_off(s, 0);
-		fs->kill_sb(s);
-		put_filesystem(fs);
-		put_super(s);
-	} else {
-		up_write(&s->s_umount);
+        if (!atomic_add_unless(&s->s_active, -1, 1)) {
+		down_write(&s->s_umount);
+		deactivate_locked_super(s);
 	}
 }
 
-EXPORT_SYMBOL(deactivate_locked_super);
+EXPORT_SYMBOL(deactivate_super);
 
 /**
  *	grab_super - acquire an active reference
@@ -247,12 +242,10 @@ static int grab_super(struct super_block *s) __releases(sb_lock)
 	/* it's going away */
 	s->s_count++;
 	spin_unlock(&sb_lock);
-	/* usually that'll be enough for it to die... */
+	/* wait for it to die */
 	down_write(&s->s_umount);
 	up_write(&s->s_umount);
 	put_super(s);
-	/* ... but in case it wasn't, let's at least yield() */
-	yield();
 	return 0;
 }