[1/2] btrfs: zoned: use rcu list in btrfs_can_activate_zone

Commit Message

Johannes Thumshirn March 7, 2022, 9:50 a.m. UTC

btrfs_can_activate_zone() can be called with the device_list_mutex already
held, which will lead to a deadlock.

As we're only traversing the list for reads we can switch from the
device_list_mutex to an rcu traversal of the list.

Fixes: a85f05e59bc1 ("btrfs: zoned: avoid chunk allocation if active block group has enough space")
Cc: Naohiro Aota <[e-mail]>
Signed-off-by: Johannes Thumshirn <[e-mail]>
---
 fs/btrfs/zoned.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Patch

diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index b7b5fac1c779..cf6341d45689 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -1986,8 +1986,8 @@  bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices, u64 flags)
 	ASSERT((flags & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0);
 
 	/* Check if there is a device with active zones left */
-	mutex_lock(&fs_devices->device_list_mutex);
-	list_for_each_entry(device, &fs_devices->devices, dev_list) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(device, &fs_devices->devices, dev_list) {
 		struct btrfs_zoned_device_info *zinfo = device->zone_info;
 
 		if (!device->bdev)
@@ -1999,7 +1999,7 @@  bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices, u64 flags)
 			break;
 		}
 	}
-	mutex_unlock(&fs_devices->device_list_mutex);
+	rcu_read_unlock();
 
 	return ret;
 }