Btrfs: Use mutex_lock_nested for tree locking

Lockdep has the notion of locking subclasses so that you can identify
locks you expect to be taken after other locks of the same class.  This
changes the per-extent buffer btree locking routines to use a subclass based
on the level in the tree.

Unfortunately, lockdep can only handle 8 total subclasses, and the btrfs
max level is also 8.  So when lockdep is on, use a lower max level.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit is contained in:
Chris Mason 2008-07-22 11:18:09 -04:00
parent f421950f86
commit 6dddcbeb28
2 changed files with 6 additions and 2 deletions

View File

@ -42,7 +42,11 @@ struct btrfs_ordered_sum;
#define BTRFS_MAGIC "_B5RfS_M"
#define BTRFS_MAX_LEVEL 8
#ifdef CONFIG_LOCKDEP
# define BTRFS_MAX_LEVEL 7
#else
# define BTRFS_MAX_LEVEL 8
#endif
/* holds pointers to all of the tree roots */
#define BTRFS_ROOT_TREE_OBJECTID 1ULL

View File

@ -37,7 +37,7 @@ int btrfs_tree_lock(struct extent_buffer *eb)
return 0;
}
cpu_relax();
mutex_lock(&eb->mutex);
mutex_lock_nested(&eb->mutex, BTRFS_MAX_LEVEL - btrfs_header_level(eb));
return 0;
}