Btrfs: make a lockdep class for the extent buffer locks

Btrfs is currently using spin_lock_nested with a nested value based
on the tree depth of the block.  But, this doesn't quite work because
the max tree depth is bigger than what spin_lock_nested can deal with,
and because locks are sometimes taken before the level field is filled in.

The solution here is to use lockdep_set_class_and_name instead, and to
set the class before unlocking the pages when the block is read from the
disk and just after init of a freshly allocated tree block.

btrfs_clear_path_blocking is also changed to take the locks in the proper
order, and it also makes sure all the locks currently held are properly
set to blocking before it tries to retake the spinlocks.  Otherwise, lockdep
gets upset about bad lock orderin.

The lockdep magic cam from Peter Zijlstra <peterz@infradead.org>

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

1 files changed, 34 insertions, 11 deletions

diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index c8f4c540cc2c..42491d728e99 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -62,14 +62,38 @@ noinline void btrfs_set_path_blocking(struct btrfs_path *p)
 
 /*
  * reset all the locked nodes in the patch to spinning locks.
+ *
+ * held is used to keep lockdep happy, when lockdep is enabled
+ * we set held to a blocking lock before we go around and
+ * retake all the spinlocks in the path.  You can safely use NULL
+ * for held
  */
-noinline void btrfs_clear_path_blocking(struct btrfs_path *p)
+noinline void btrfs_clear_path_blocking(struct btrfs_path *p,
+                                        struct extent_buffer *held)
 {
         int i;
-        for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+        /* lockdep really cares that we take all of these spinlocks
+         * in the right order.  If any of the locks in the path are not
+         * currently blocking, it is going to complain.  So, make really
+         * really sure by forcing the path to blocking before we clear
+         * the path blocking.
+         */
+        if (held)
+                btrfs_set_lock_blocking(held);
+        btrfs_set_path_blocking(p);
+#endif
+
+        for (i = BTRFS_MAX_LEVEL - 1; i >= 0; i--) {
                 if (p->nodes[i] && p->locks[i])
                         btrfs_clear_lock_blocking(p->nodes[i]);
         }
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+        if (held)
+                btrfs_clear_lock_blocking(held);
+#endif
 }
 
 /* this also releases the path */
@@ -279,7 +303,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
                                                   trans->transid, level, &ins);
                 BUG_ON(ret);
                 cow = btrfs_init_new_buffer(trans, root, prealloc_dest,
-                                            buf->len);
+                                            buf->len, level);
         } else {
                 cow = btrfs_alloc_free_block(trans, root, buf->len,
                                              parent_start,
@@ -1559,7 +1583,7 @@ cow_done:
                 if (!p->skip_locking)
                         p->locks[level] = 1;
 
-                btrfs_clear_path_blocking(p);
+                btrfs_clear_path_blocking(p, NULL);
 
                 /*
                  * we have a lock on b and as long as we aren't changing
@@ -1598,7 +1622,7 @@ cow_done:
 
                                 btrfs_set_path_blocking(p);
                                 sret = split_node(trans, root, p, level);
-                                btrfs_clear_path_blocking(p);
+                                btrfs_clear_path_blocking(p, NULL);
 
                                 BUG_ON(sret > 0);
                                 if (sret) {
@@ -1618,7 +1642,7 @@ cow_done:
 
                                 btrfs_set_path_blocking(p);
                                 sret = balance_level(trans, root, p, level);
-                                btrfs_clear_path_blocking(p);
+                                btrfs_clear_path_blocking(p, NULL);
 
                                 if (sret) {
                                         ret = sret;
@@ -1681,13 +1705,13 @@ cow_done:
                         if (!p->skip_locking) {
                                 int lret;
 
-                                btrfs_clear_path_blocking(p);
+                                btrfs_clear_path_blocking(p, NULL);
                                 lret = btrfs_try_spin_lock(b);
 
                                 if (!lret) {
                                         btrfs_set_path_blocking(p);
                                         btrfs_tree_lock(b);
-                                        btrfs_clear_path_blocking(p);
+                                        btrfs_clear_path_blocking(p, b);
                                 }
                         }
                 } else {
@@ -1699,7 +1723,7 @@ cow_done:
                                 btrfs_set_path_blocking(p);
                                 sret = split_leaf(trans, root, key,
                                                       p, ins_len, ret == 0);
-                                btrfs_clear_path_blocking(p);
+                                btrfs_clear_path_blocking(p, NULL);
 
                                 BUG_ON(sret > 0);
                                 if (sret) {
@@ -3919,7 +3943,6 @@ find_next_key:
                                 btrfs_release_path(root, path);
                                 goto again;
                         } else {
-                                btrfs_clear_path_blocking(path);
                                 goto out;
                         }
                 }
@@ -3939,7 +3962,7 @@ find_next_key:
                 path->locks[level - 1] = 1;
                 path->nodes[level - 1] = cur;
                 unlock_up(path, level, 1);
-                btrfs_clear_path_blocking(path);
+                btrfs_clear_path_blocking(path, NULL);
         }
 out:
         if (ret == 0)