aboutsummaryrefslogtreecommitdiff
path: root/fs/btrfs/ordered-data.c
diff options
context:
space:
mode:
authorChris Mason <chris.mason@oracle.com>2008-07-24 11:57:52 -0400
committerChris Mason <chris.mason@oracle.com>2008-09-25 11:04:05 -0400
commit3eaa2885276fd6dac7b076a793932428b7168e74 (patch)
treeb06382bec68bf1755597a74ac8225f3bcddda5e5 /fs/btrfs/ordered-data.c
parent64f26f745084872b916cd1bef6054e21b15c5784 (diff)
Btrfs: Fix the defragmention code and the block relocation code for data=ordered
Before setting an extent to delalloc, the code needs to wait for pending ordered extents. Also, the relocation code needs to wait for ordered IO before scanning the block group again. This is because the extents are not removed until the IO for the new extents is finished Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs/btrfs/ordered-data.c')
-rw-r--r--fs/btrfs/ordered-data.c56
1 files changed, 54 insertions, 2 deletions
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index e42fd233e04..676e4bd65c5 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -167,20 +167,28 @@ int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
entry->file_offset = file_offset;
entry->start = start;
entry->len = len;
+ entry->inode = inode;
+
/* one ref for the tree */
atomic_set(&entry->refs, 1);
init_waitqueue_head(&entry->wait);
INIT_LIST_HEAD(&entry->list);
+ INIT_LIST_HEAD(&entry->root_extent_list);
node = tree_insert(&tree->tree, file_offset,
&entry->rb_node);
if (node) {
- entry = rb_entry(node, struct btrfs_ordered_extent, rb_node);
- atomic_inc(&entry->refs);
+ printk("warning dup entry from add_ordered_extent\n");
+ BUG();
}
set_extent_ordered(&BTRFS_I(inode)->io_tree, file_offset,
entry_end(entry) - 1, GFP_NOFS);
+ spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+ list_add_tail(&entry->root_extent_list,
+ &BTRFS_I(inode)->root->fs_info->ordered_extents);
+ spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+
mutex_unlock(&tree->mutex);
BUG_ON(node);
return 0;
@@ -285,11 +293,55 @@ int btrfs_remove_ordered_extent(struct inode *inode,
rb_erase(node, &tree->tree);
tree->last = NULL;
set_bit(BTRFS_ORDERED_COMPLETE, &entry->flags);
+
+ spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+ list_del_init(&entry->root_extent_list);
+ spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
+
mutex_unlock(&tree->mutex);
wake_up(&entry->wait);
return 0;
}
+int btrfs_wait_ordered_extents(struct btrfs_root *root)
+{
+ struct list_head splice;
+ struct list_head *cur;
+ struct btrfs_ordered_extent *ordered;
+ struct inode *inode;
+
+ INIT_LIST_HEAD(&splice);
+
+ spin_lock(&root->fs_info->ordered_extent_lock);
+ list_splice_init(&root->fs_info->ordered_extents, &splice);
+ while(!list_empty(&splice)) {
+ cur = splice.next;
+ ordered = list_entry(cur, struct btrfs_ordered_extent,
+ root_extent_list);
+ list_del_init(&ordered->root_extent_list);
+ atomic_inc(&ordered->refs);
+ inode = ordered->inode;
+
+ /*
+ * the inode can't go away until all the pages are gone
+ * and the pages won't go away while there is still
+ * an ordered extent and the ordered extent won't go
+ * away until it is off this list. So, we can safely
+ * increment i_count here and call iput later
+ */
+ atomic_inc(&inode->i_count);
+ spin_unlock(&root->fs_info->ordered_extent_lock);
+
+ btrfs_start_ordered_extent(inode, ordered, 1);
+ btrfs_put_ordered_extent(ordered);
+ iput(inode);
+
+ spin_lock(&root->fs_info->ordered_extent_lock);
+ }
+ spin_unlock(&root->fs_info->ordered_extent_lock);
+ return 0;
+}
+
/*
* Used to start IO or wait for a given ordered extent to finish.
*