ocfs2: Abstract ocfs2_extent_tree in b-tree operations.

In the old extent tree operation, we take the hypothesis that we
are using the ocfs2_extent_list in ocfs2_dinode as the tree root.
As xattr will also use ocfs2_extent_list to store large value
for a xattr entry, we refactor the tree operation so that xattr
can use it directly.

The refactoring includes 4 steps:
1. Abstract set/get of last_eb_blk and update_clusters since they may
   be stored in different location for dinode and xattr.
2. Add a new structure named ocfs2_extent_tree to indicate the
   extent tree the operation will work on.
3. Remove all the use of fe_bh and di, use root_bh and root_el in
   extent tree instead. So now all the fe_bh is replaced with
   et->root_bh, el with root_el accordingly.
4. Make ocfs2_lock_allocators generic. Now it is limited to be only used
   in file extend allocation. But the whole function is useful when we want
   to store large EAs.

Note: This patch doesn't touch ocfs2_commit_truncate() since it is not used
for anything other than truncate inode data btrees.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>

1 files changed, 82 insertions, 0 deletions

diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index 2a817bca1ddb..b642c825fb7c 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -1894,3 +1894,85 @@ static inline void ocfs2_debug_suballoc_inode(struct ocfs2_dinode *fe)
                        (unsigned long long)fe->id2.i_chain.cl_recs[i].c_blkno);
         }
 }
+
+/*
+ * For a given allocation, determine which allocators will need to be
+ * accessed, and lock them, reserving the appropriate number of bits.
+ *
+ * Sparse file systems call this from ocfs2_write_begin_nolock()
+ * and ocfs2_allocate_unwritten_extents().
+ *
+ * File systems which don't support holes call this from
+ * ocfs2_extend_allocation().
+ */
+int ocfs2_lock_allocators(struct inode *inode, struct buffer_head *root_bh,
+                          struct ocfs2_extent_list *root_el,
+                          u32 clusters_to_add, u32 extents_to_split,
+                          struct ocfs2_alloc_context **data_ac,
+                          struct ocfs2_alloc_context **meta_ac)
+{
+        int ret = 0, num_free_extents;
+        unsigned int max_recs_needed = clusters_to_add + 2 * extents_to_split;
+        struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+        *meta_ac = NULL;
+        if (data_ac)
+                *data_ac = NULL;
+
+        BUG_ON(clusters_to_add != 0 && data_ac == NULL);
+
+        num_free_extents = ocfs2_num_free_extents(osb, inode, root_bh,
+                                                  OCFS2_DINODE_EXTENT);
+        if (num_free_extents < 0) {
+                ret = num_free_extents;
+                mlog_errno(ret);
+                goto out;
+        }
+
+        /*
+         * Sparse allocation file systems need to be more conservative
+         * with reserving room for expansion - the actual allocation
+         * happens while we've got a journal handle open so re-taking
+         * a cluster lock (because we ran out of room for another
+         * extent) will violate ordering rules.
+         *
+         * Most of the time we'll only be seeing this 1 cluster at a time
+         * anyway.
+         *
+         * Always lock for any unwritten extents - we might want to
+         * add blocks during a split.
+         */
+        if (!num_free_extents ||
+            (ocfs2_sparse_alloc(osb) && num_free_extents < max_recs_needed)) {
+                ret = ocfs2_reserve_new_metadata(osb, root_el, meta_ac);
+                if (ret < 0) {
+                        if (ret != -ENOSPC)
+                                mlog_errno(ret);
+                        goto out;
+                }
+        }
+
+        if (clusters_to_add == 0)
+                goto out;
+
+        ret = ocfs2_reserve_clusters(osb, clusters_to_add, data_ac);
+        if (ret < 0) {
+                if (ret != -ENOSPC)
+                        mlog_errno(ret);
+                goto out;
+        }
+
+out:
+        if (ret) {
+                if (*meta_ac) {
+                        ocfs2_free_alloc_context(*meta_ac);
+                        *meta_ac = NULL;
+                }
+
+                /*
+                 * We cannot have an error and a non null *data_ac.
+                 */
+        }
+
+        return ret;
+}