Btrfs: fix race between removing a dev and writing sbs
This change fixes an issue when removing a device and writing
all super blocks run simultaneously. Here's the steps necessary
for the issue to happen:
1) disk-io.c:write_all_supers() gets a number of N devices from the
super_copy, so it will not panic if it fails to write super blocks
for N - 1 devices;
2) Then it tries to acquire the device_list_mutex, but blocks because
volumes.c:btrfs_rm_device() got it first;
3) btrfs_rm_device() removes the device from the list, then unlocks the
mutex and after the unlock it updates the number of devices in
super_copy to N - 1.
4) write_all_supers() finally acquires the mutex, iterates over all the
devices in the list and gets N - 1 errors, that is, it failed to write
super blocks to all the devices;
5) Because write_all_supers() thinks there are a total of N devices, it
considers N - 1 errors to be ok, and therefore won't panic.
So this change just makes sure that write_all_supers() reads the number
of devices from super_copy after it acquires the device_list_mutex.
Conversely, it changes btrfs_rm_device() to update the number of devices
in super_copy before it releases the device list mutex.
The code path to add a new device (volumes.c:btrfs_init_new_device),
already has the right behaviour: it updates the number of devices in
super_copy while holding the device_list_mutex.
The only code path that doesn't lock the device list mutex
before updating the number of devices in the super copy is
disk-io.c:next_root_backup(), called by open_ctree() during
mount time where concurrency issues can't happen.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 940bc48..6d642e4 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -3365,7 +3365,6 @@
int total_errors = 0;
u64 flags;
- max_errors = btrfs_super_num_devices(root->fs_info->super_copy) - 1;
do_barriers = !btrfs_test_opt(root, NOBARRIER);
backup_super_roots(root->fs_info);
@@ -3374,6 +3373,7 @@
mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
head = &root->fs_info->fs_devices->devices;
+ max_errors = btrfs_super_num_devices(root->fs_info->super_copy) - 1;
if (do_barriers) {
ret = barrier_all_devices(root->fs_info);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 34068b8..3f1c2c2 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1620,7 +1620,11 @@
/*
* the device list mutex makes sure that we don't change
* the device list while someone else is writing out all
- * the device supers.
+ * the device supers. Whoever is writing all supers, should
+ * lock the device list mutex before getting the number of
+ * devices in the super block (super_copy). Conversely,
+ * whoever updates the number of devices in the super block
+ * (super_copy) should hold the device list mutex.
*/
cur_devices = device->fs_devices;
@@ -1644,10 +1648,10 @@
device->fs_devices->open_devices--;
call_rcu(&device->rcu, free_device);
- mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
num_devices = btrfs_super_num_devices(root->fs_info->super_copy) - 1;
btrfs_set_super_num_devices(root->fs_info->super_copy, num_devices);
+ mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
if (cur_devices->open_devices == 0) {
struct btrfs_fs_devices *fs_devices;