[PATCH v4] nilfs2: fix checkpoint root lifetime on sysfs errors

[Aldo Ariel Panzardo]

nilfs_find_or_create_root() links a new checkpoint root into the
checkpoint tree and drops ns_cptree_lock before creating the root's
sysfs group.  If nilfs_sysfs_create_snapshot_group() then fails, the
root is freed while it is still reachable from ns_cptree, so a
concurrent nilfs_lookup_root() can dereference freed memory.

nilfs_sysfs_create_snapshot_group() allocates memory and creates sysfs
nodes and may sleep, so it cannot run under the ns_cptree_lock spinlock.
Reorder the creation path so that the sysfs group is set up before the
root is published: initialize the root, create its sysfs group outside
the lock, and only then take ns_cptree_lock to link it into the tree.
On sysfs failure the root was never visible and is freed directly; in
the unlikely case a root with the same checkpoint number is already
present, the sysfs group of the new root is removed before it is freed.
Concurrent insertions are already serialized by ns_snapshot_mount_mutex,
so only the race against nilfs_lookup_root() needs to be closed.

Because kobject_put() does not guarantee that the embedded kobject's
release callback has run by the time it returns
(CONFIG_DEBUG_KOBJECT_RELEASE defers it), the container must not be
freed until the release completes.  Keep that wait_for_completion() of
snapshot_kobj_unregister inside the sysfs helpers, so both the creation
error path and nilfs_sysfs_delete_snapshot_group() wait for the release
before their callers free the root.

Fixes: dd70edbde262 ("nilfs2: integrate sysfs support into driver")
Cc: stable@xxxxxxxxxxxxxxx
Signed-off-by: Aldo Ariel Panzardo <qwe.aldo@xxxxxxxxx>
---
Changes in v4:
  - Keep ns_cptree_lock as a spinlock instead of converting it to a
    mutex, and create the sysfs group before taking the lock, as
    suggested by Ryusuke Konishi. Concurrent insertions are already
    serialized by ns_snapshot_mount_mutex; only the race against
    nilfs_lookup_root() needs to be closed.
  - Move the wait_for_completion() of snapshot_kobj_unregister into the
    sysfs helpers (create error path and delete), so the checkpoint
    tree functions no longer manage the kobject completion state, as
    suggested by Ryusuke Konishi.

Changes in v3:
  - Move wait_for_completion() out of the locked section in
    nilfs_put_root() (the mutex-based v2/v3 approach; dropped in v4).

Changes in v2:
  - Serialize the sysfs registration against the tree instead of adding
    a second lock, as suggested by Viacheslav Dubeyko.

 fs/nilfs2/sysfs.c     |  5 ++++-
 fs/nilfs2/the_nilfs.c | 34 +++++++++++++++++++++-------------
 2 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/fs/nilfs2/sysfs.c b/fs/nilfs2/sysfs.c
index bc52afbfc5c7..5bb74a45ad1e 100644
--- a/fs/nilfs2/sysfs.c
+++ b/fs/nilfs2/sysfs.c
@@ -195,8 +195,10 @@ int nilfs_sysfs_create_snapshot_group(struct nilfs_root *root)
 					    "%llu", root->cno);
 	}
 
-	if (err)
+	if (err) {
 		kobject_put(&root->snapshot_kobj);
+		wait_for_completion(&root->snapshot_kobj_unregister);
+	}
 
 	return err;
 }
@@ -204,6 +206,7 @@ int nilfs_sysfs_create_snapshot_group(struct nilfs_root *root)
 void nilfs_sysfs_delete_snapshot_group(struct nilfs_root *root)
 {
 	kobject_put(&root->snapshot_kobj);
+	wait_for_completion(&root->snapshot_kobj_unregister);
 }
 
 /************************************************************************
diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c
index ecd71c190885..1d5d2293c9c7 100644
--- a/fs/nilfs2/the_nilfs.c
+++ b/fs/nilfs2/the_nilfs.c
@@ -869,6 +869,26 @@ nilfs_find_or_create_root(struct the_nilfs *nilfs, __u64 cno)
 	if (!new)
 		return NULL;
 
+	new->cno = cno;
+	new->ifile = NULL;
+	new->nilfs = nilfs;
+	refcount_set(&new->count, 1);
+	atomic64_set(&new->inodes_count, 0);
+	atomic64_set(&new->blocks_count, 0);
+
+	/*
+	 * Register the sysfs group before publishing the root in the
+	 * checkpoint tree.  nilfs_sysfs_create_snapshot_group() can sleep,
+	 * so it must run outside ns_cptree_lock; creating it first also
+	 * ensures a concurrent nilfs_lookup_root() can never observe a root
+	 * whose sysfs registration later fails and gets freed.
+	 */
+	err = nilfs_sysfs_create_snapshot_group(new);
+	if (err) {
+		kfree(new);
+		return NULL;
+	}
+
 	spin_lock(&nilfs->ns_cptree_lock);
 
 	p = &nilfs->ns_cptree.rb_node;
@@ -885,29 +905,17 @@ nilfs_find_or_create_root(struct the_nilfs *nilfs, __u64 cno)
 		} else {
 			refcount_inc(&root->count);
 			spin_unlock(&nilfs->ns_cptree_lock);
+			nilfs_sysfs_delete_snapshot_group(new);
 			kfree(new);
 			return root;
 		}
 	}
 
-	new->cno = cno;
-	new->ifile = NULL;
-	new->nilfs = nilfs;
-	refcount_set(&new->count, 1);
-	atomic64_set(&new->inodes_count, 0);
-	atomic64_set(&new->blocks_count, 0);
-
 	rb_link_node(&new->rb_node, parent, p);
 	rb_insert_color(&new->rb_node, &nilfs->ns_cptree);
 
 	spin_unlock(&nilfs->ns_cptree_lock);
 
-	err = nilfs_sysfs_create_snapshot_group(new);
-	if (err) {
-		kfree(new);
-		new = NULL;
-	}
-
 	return new;
 }
 
-- 
2.43.0