Btrfs: avoid races between super writeout and device list updates

On multi-device filesystems, btrfs writes supers to all of the devices
before considering a sync complete.  There wasn't any additional
locking between super writeout and the device list management code
because device management was done inside a transaction and
super writeout only happened  with no transation writers running.

With the btrfs fsync log and other async transaction updates, this
has been racey for some time.  This adds a mutex to protect
the device list.  The existing volume mutex could not be reused due to
transaction lock ordering requirements.

Signed-off-by: Chris Mason <chris.mason@oracle.com>

1 files changed, 5 insertions, 1 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 6c54c210dfd0..b7ddc77fa568 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -2111,7 +2111,7 @@ static int write_dev_supers(struct btrfs_device *device,
 
 int write_all_supers(struct btrfs_root *root, int max_mirrors)
 {
-	struct list_head *head = &root->fs_info->fs_devices->devices;
+	struct list_head *head;
 	struct btrfs_device *dev;
 	struct btrfs_super_block *sb;
 	struct btrfs_dev_item *dev_item;
@@ -2126,6 +2126,9 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors)
 
 	sb = &root->fs_info->super_for_commit;
 	dev_item = &sb->dev_item;
+
+	mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
+	head = &root->fs_info->fs_devices->devices;
 	list_for_each_entry(dev, head, dev_list) {
 		if (!dev->bdev) {
 			total_errors++;
@@ -2169,6 +2172,7 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors)
 		if (ret)
 			total_errors++;
 	}
+	mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
 	if (total_errors > max_errors) {
 		printk(KERN_ERR "btrfs: %d errors while writing supers\n",
 		       total_errors);