btrfs: allocate raid type kobjects dynamically

commit c1895442be01c58449e3bf9272f22062a670e08f upstream.

We are currently allocating space_info objects in an array when we
allocate space_info. When a user does something like:

# btrfs balance start -mconvert=raid1 -dconvert=raid1 /mnt
# btrfs balance start -mconvert=single -dconvert=single /mnt -f
# btrfs balance start -mconvert=raid1 -dconvert=raid1 /

We can end up with memory corruption since the kobject hasn't
been reinitialized properly and the name pointer was left set.

The rationale behind allocating them statically was to avoid
creating a separate kobject container that just contained the
raid type. It used the index in the array to determine the index.

Ultimately, though, this wastes more memory than it saves in all
but the most complex scenarios and introduces kobject lifetime
questions.

This patch allocates the kobjects dynamically instead. Note that
we also remove the kobject_get/put of the parent kobject since
kobject_add and kobject_del do that internally.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reported-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

3 files changed, 36 insertions, 16 deletions

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2c1a42ca519f..d3511cc17091 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1104,6 +1104,12 @@ struct btrfs_qgroup_limit_item {
 	__le64 rsv_excl;
 } __attribute__ ((__packed__));
 
+/* For raid type sysfs entries */
+struct raid_kobject {
+	int raid_type;
+	struct kobject kobj;
+};
+
 struct btrfs_space_info {
 	spinlock_t lock;
 
@@ -1154,7 +1160,7 @@ struct btrfs_space_info {
 	wait_queue_head_t wait;
 
 	struct kobject kobj;
-	struct kobject block_group_kobjs[BTRFS_NR_RAID_TYPES];
+	struct kobject *block_group_kobjs[BTRFS_NR_RAID_TYPES];
 };
 
 #define	BTRFS_BLOCK_RSV_GLOBAL		1
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 1bbb204c8a5d..3ff98e23f651 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3400,10 +3400,8 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 		return ret;
 	}
 
-	for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
+	for (i = 0; i < BTRFS_NR_RAID_TYPES; i++)
 		INIT_LIST_HEAD(&found->block_groups[i]);
-		kobject_init(&found->block_group_kobjs[i], &btrfs_raid_ktype);
-	}
 	init_rwsem(&found->groups_sem);
 	spin_lock_init(&found->lock);
 	found->flags = flags & BTRFS_BLOCK_GROUP_TYPE_MASK;
@@ -8328,8 +8326,9 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		list_del(&space_info->list);
 		for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
 			struct kobject *kobj;
-			kobj = &space_info->block_group_kobjs[i];
-			if (kobj->parent) {
+			kobj = space_info->block_group_kobjs[i];
+			space_info->block_group_kobjs[i] = NULL;
+			if (kobj) {
 				kobject_del(kobj);
 				kobject_put(kobj);
 			}
@@ -8353,17 +8352,26 @@ static void __link_block_group(struct btrfs_space_info *space_info,
 	up_write(&space_info->groups_sem);
 
 	if (first) {
-		struct kobject *kobj = &space_info->block_group_kobjs[index];
+		struct raid_kobject *rkobj;
 		int ret;
 
-		kobject_get(&space_info->kobj); /* put in release */
-		ret = kobject_add(kobj, &space_info->kobj, "%s",
-				  get_raid_name(index));
+		rkobj = kzalloc(sizeof(*rkobj), GFP_NOFS);
+		if (!rkobj)
+			goto out_err;
+		rkobj->raid_type = index;
+		kobject_init(&rkobj->kobj, &btrfs_raid_ktype);
+		ret = kobject_add(&rkobj->kobj, &space_info->kobj,
+				  "%s", get_raid_name(index));
 		if (ret) {
-			pr_warn("BTRFS: failed to add kobject for block cache. ignoring.\n");
-			kobject_put(&space_info->kobj);
+			kobject_put(&rkobj->kobj);
+			goto out_err;
 		}
+		space_info->block_group_kobjs[index] = &rkobj->kobj;
 	}
+
+	return;
+out_err:
+	pr_warn("BTRFS: failed to add kobject for block cache. ignoring.\n");
 }
 
 static struct btrfs_block_group_cache *
@@ -8698,6 +8706,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	struct btrfs_root *tree_root = root->fs_info->tree_root;
 	struct btrfs_key key;
 	struct inode *inode;
+	struct kobject *kobj = NULL;
 	int ret;
 	int index;
 	int factor;
@@ -8797,11 +8806,15 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	 */
 	list_del_init(&block_group->list);
 	if (list_empty(&block_group->space_info->block_groups[index])) {
-		kobject_del(&block_group->space_info->block_group_kobjs[index]);
-		kobject_put(&block_group->space_info->block_group_kobjs[index]);
+		kobj = block_group->space_info->block_group_kobjs[index];
+		block_group->space_info->block_group_kobjs[index] = NULL;
 		clear_avail_alloc_bits(root->fs_info, block_group->flags);
 	}
 	up_write(&block_group->space_info->groups_sem);
+	if (kobj) {
+		kobject_del(kobj);
+		kobject_put(kobj);
+	}
 
 	if (block_group->cached == BTRFS_CACHE_STARTED)
 		wait_block_group_cache_done(block_group);
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 865f4cf9a769..ff286f372ff4 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -253,6 +253,7 @@ static ssize_t global_rsv_reserved_show(struct kobject *kobj,
 BTRFS_ATTR(global_rsv_reserved, 0444, global_rsv_reserved_show);
 
 #define to_space_info(_kobj) container_of(_kobj, struct btrfs_space_info, kobj)
+#define to_raid_kobj(_kobj) container_of(_kobj, struct raid_kobject, kobj)
 
 static ssize_t raid_bytes_show(struct kobject *kobj,
 			       struct kobj_attribute *attr, char *buf);
@@ -265,7 +266,7 @@ static ssize_t raid_bytes_show(struct kobject *kobj,
 {
 	struct btrfs_space_info *sinfo = to_space_info(kobj->parent);
 	struct btrfs_block_group_cache *block_group;
-	int index = kobj - sinfo->block_group_kobjs;
+	int index = to_raid_kobj(kobj)->raid_type;
 	u64 val = 0;
 
 	down_read(&sinfo->groups_sem);
@@ -287,7 +288,7 @@ static struct attribute *raid_attributes[] = {
 
 static void release_raid_kobj(struct kobject *kobj)
 {
-	kobject_put(kobj->parent);
+	kfree(to_raid_kobj(kobj));
 }
 
 struct kobj_type btrfs_raid_ktype = {