Btrfs: fix that check_int_data mount option was ignored
authorStefan Behrens <sbehrens@giantdisaster.de>
Fri, 30 Mar 2012 11:58:32 +0000 (13:58 +0200)
committerDavid Sterba <dsterba@suse.cz>
Wed, 18 Apr 2012 17:22:38 +0000 (19:22 +0200)
The bitfield member mount_opt was too small by one bit to hold the mount
option that enabled to include data extents in the integrity checker.
Since the same issue happened when the BTRFS_MOUNT_PANIC_ON_FATAL_ERROR
option was added (git rebase silently merges so that the increase of the
size of the bitfield member is lost), the bit limit was removed entirely.

Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
fs/btrfs/ctree.h

index 5b8ef8eb35218e0f595ee93069fc1ccab49c6b1e..ec42a24e935e8345c771ece0a40478d6e01aa669 100644 (file)
@@ -1078,7 +1078,7 @@ struct btrfs_fs_info {
         * is required instead of the faster short fsync log commits
         */
        u64 last_trans_log_full_commit;
-       unsigned long mount_opt:21;
+       unsigned long mount_opt;
        unsigned long compress_type:4;
        u64 max_inline;
        u64 alloc_start;