4 files changed, 148 insertions, 18 deletions

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index bc57e236ac64..2b15fb97d23f 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2006,6 +2006,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
 int btrfs_extent_readonly(struct btrfs_root *root, u64 bytenr);
 int btrfs_free_block_groups(struct btrfs_fs_info *info);
 int btrfs_read_block_groups(struct btrfs_root *root);
+int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr);
 int btrfs_make_block_group(struct btrfs_trans_handle *trans,
                            struct btrfs_root *root, u64 bytes_used,
                            u64 type, u64 chunk_objectid, u64 chunk_offset,
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 4bd04f3fa8bb..4c7c9467f224 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -7402,6 +7402,93 @@ out:
 }
 #endif
 
+/*
+ * checks to see if its even possible to relocate this block group.
+ *
+ * @return - -1 if it's not a good idea to relocate this block group, 0 if its
+ * ok to go ahead and try.
+ */
+int btrfs_can_relocate(struct btrfs_root *root, u64 bytenr)
+{
+        struct btrfs_block_group_cache *block_group;
+        struct btrfs_space_info *space_info;
+        struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
+        struct btrfs_device *device;
+        int full = 0;
+        int ret = 0;
+
+        block_group = btrfs_lookup_block_group(root->fs_info, bytenr);
+
+        /* odd, couldn't find the block group, leave it alone */
+        if (!block_group)
+                return -1;
+
+        /* no bytes used, we're good */
+        if (!btrfs_block_group_used(&block_group->item))
+                goto out;
+
+        space_info = block_group->space_info;
+        spin_lock(&space_info->lock);
+
+        full = space_info->full;
+
+        /*
+         * if this is the last block group we have in this space, we can't
+         * relocate it.
+         */
+        if (space_info->total_bytes == block_group->key.offset) {
+                ret = -1;
+                spin_unlock(&space_info->lock);
+                goto out;
+        }
+
+        /*
+         * need to make sure we have room in the space to handle all of the
+         * extents from this block group.  If we can, we're good
+         */
+        if (space_info->bytes_used + space_info->bytes_reserved +
+            space_info->bytes_pinned + space_info->bytes_readonly +
+            btrfs_block_group_used(&block_group->item) <
+            space_info->total_bytes) {
+                spin_unlock(&space_info->lock);
+                goto out;
+        }
+        spin_unlock(&space_info->lock);
+
+        /*
+         * ok we don't have enough space, but maybe we have free space on our
+         * devices to allocate new chunks for relocation, so loop through our
+         * alloc devices and guess if we have enough space.  However, if we
+         * were marked as full, then we know there aren't enough chunks, and we
+         * can just return.
+         */
+        ret = -1;
+        if (full)
+                goto out;
+
+        mutex_lock(&root->fs_info->chunk_mutex);
+        list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
+                u64 min_free = btrfs_block_group_used(&block_group->item);
+                u64 dev_offset, max_avail;
+
+                /*
+                 * check to make sure we can actually find a chunk with enough
+                 * space to fit our block group in.
+                 */
+                if (device->total_bytes > device->bytes_used + min_free) {
+                        ret = find_free_dev_extent(NULL, device, min_free,
+                                                   &dev_offset, &max_avail);
+                        if (!ret)
+                                break;
+                        ret = -1;
+                }
+        }
+        mutex_unlock(&root->fs_info->chunk_mutex);
+out:
+        btrfs_put_block_group(block_group);
+        return ret;
+}
+
 static int find_first_block_group(struct btrfs_root *root,
                 struct btrfs_path *path, struct btrfs_key *key)
 {
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index d2358c06bbd9..be953afe804c 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -719,10 +719,9 @@ error:
  * called very infrequently and that a given device has a small number
  * of extents
  */
-static noinline int find_free_dev_extent(struct btrfs_trans_handle *trans,
-                                         struct btrfs_device *device,
-                                         u64 num_bytes, u64 *start,
-                                         u64 *max_avail)
+int find_free_dev_extent(struct btrfs_trans_handle *trans,
+                         struct btrfs_device *device, u64 num_bytes,
+                         u64 *start, u64 *max_avail)
 {
         struct btrfs_key key;
         struct btrfs_root *root = device->dev_root;
@@ -1736,6 +1735,10 @@ static int btrfs_relocate_chunk(struct btrfs_root *root,
         extent_root = root->fs_info->extent_root;
         em_tree = &root->fs_info->mapping_tree.map_tree;
 
+        ret = btrfs_can_relocate(extent_root, chunk_offset);
+        if (ret)
+                return -ENOSPC;
+
         /* step one, relocate all the extents inside this chunk */
         ret = btrfs_relocate_block_group(extent_root, chunk_offset);
         BUG_ON(ret);
@@ -1807,12 +1810,15 @@ static int btrfs_relocate_sys_chunks(struct btrfs_root *root)
         struct btrfs_key found_key;
         u64 chunk_tree = chunk_root->root_key.objectid;
         u64 chunk_type;
+        bool retried = false;
+        int failed = 0;
         int ret;
 
         path = btrfs_alloc_path();
         if (!path)
                 return -ENOMEM;
 
+again:
         key.objectid = BTRFS_FIRST_CHUNK_TREE_OBJECTID;
         key.offset = (u64)-1;
         key.type = BTRFS_CHUNK_ITEM_KEY;
@@ -1842,7 +1848,10 @@ static int btrfs_relocate_sys_chunks(struct btrfs_root *root)
                         ret = btrfs_relocate_chunk(chunk_root, chunk_tree,
                                                    found_key.objectid,
                                                    found_key.offset);
-                        BUG_ON(ret);
+                        if (ret == -ENOSPC)
+                                failed++;
+                        else if (ret)
+                                BUG();
                 }
 
                 if (found_key.offset == 0)
@@ -1850,6 +1859,14 @@ static int btrfs_relocate_sys_chunks(struct btrfs_root *root)
                 key.offset = found_key.offset - 1;
         }
         ret = 0;
+        if (failed && !retried) {
+                failed = 0;
+                retried = true;
+                goto again;
+        } else if (failed && retried) {
+                WARN_ON(1);
+                ret = -ENOSPC;
+        }
 error:
         btrfs_free_path(path);
         return ret;
@@ -1894,6 +1911,8 @@ int btrfs_balance(struct btrfs_root *dev_root)
                         continue;
 
                 ret = btrfs_shrink_device(device, old_size - size_to_free);
+                if (ret == -ENOSPC)
+                        break;
                 BUG_ON(ret);
 
                 trans = btrfs_start_transaction(dev_root, 1);
@@ -1938,9 +1957,8 @@ int btrfs_balance(struct btrfs_root *dev_root)
                 chunk = btrfs_item_ptr(path->nodes[0],
                                        path->slots[0],
                                        struct btrfs_chunk);
-                key.offset = found_key.offset;
                 /* chunk zero is special */
-                if (key.offset == 0)
+                if (found_key.offset == 0)
                         break;
 
                 btrfs_release_path(chunk_root, path);
@@ -1948,7 +1966,8 @@ int btrfs_balance(struct btrfs_root *dev_root)
                                            chunk_root->root_key.objectid,
                                            found_key.objectid,
                                            found_key.offset);
-                BUG_ON(ret);
+                BUG_ON(ret && ret != -ENOSPC);
+                key.offset = found_key.offset - 1;
         }
         ret = 0;
 error:
@@ -1974,10 +1993,13 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
         u64 chunk_offset;
         int ret;
         int slot;
+        int failed = 0;
+        bool retried = false;
         struct extent_buffer *l;
         struct btrfs_key key;
         struct btrfs_super_block *super_copy = &root->fs_info->super_copy;
         u64 old_total = btrfs_super_total_bytes(super_copy);
+        u64 old_size = device->total_bytes;
         u64 diff = device->total_bytes - new_size;
 
         if (new_size >= device->total_bytes)
@@ -1987,12 +2009,6 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
         if (!path)
                 return -ENOMEM;
 
-        trans = btrfs_start_transaction(root, 1);
-        if (!trans) {
-                ret = -ENOMEM;
-                goto done;
-        }
-
         path->reada = 2;
 
         lock_chunks(root);
@@ -2001,8 +2017,8 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
         if (device->writeable)
                 device->fs_devices->total_rw_bytes -= diff;
         unlock_chunks(root);
-        btrfs_end_transaction(trans, root);
 
+again:
         key.objectid = device->devid;
         key.offset = (u64)-1;
         key.type = BTRFS_DEV_EXTENT_KEY;
@@ -2017,6 +2033,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
                         goto done;
                 if (ret) {
                         ret = 0;
+                        btrfs_release_path(root, path);
                         break;
                 }
 
@@ -2024,14 +2041,18 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
                 slot = path->slots[0];
                 btrfs_item_key_to_cpu(l, &key, path->slots[0]);
 
-                if (key.objectid != device->devid)
+                if (key.objectid != device->devid) {
+                        btrfs_release_path(root, path);
                         break;
+                }
 
                 dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent);
                 length = btrfs_dev_extent_length(l, dev_extent);
 
-                if (key.offset + length <= new_size)
+                if (key.offset + length <= new_size) {
+                        btrfs_release_path(root, path);
                         break;
+                }
 
                 chunk_tree = btrfs_dev_extent_chunk_tree(l, dev_extent);
                 chunk_objectid = btrfs_dev_extent_chunk_objectid(l, dev_extent);
@@ -2040,8 +2061,26 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size)
 
                 ret = btrfs_relocate_chunk(root, chunk_tree, chunk_objectid,
                                            chunk_offset);
-                if (ret)
+                if (ret && ret != -ENOSPC)
                         goto done;
+                if (ret == -ENOSPC)
+                        failed++;
+                key.offset -= 1;
+        }
+
+        if (failed && !retried) {
+                failed = 0;
+                retried = true;
+                goto again;
+        } else if (failed && retried) {
+                ret = -ENOSPC;
+                lock_chunks(root);
+
+                device->total_bytes = old_size;
+                if (device->writeable)
+                        device->fs_devices->total_rw_bytes += diff;
+                unlock_chunks(root);
+                goto done;
         }
 
         /* Shrinking succeeded, else we would be at "done". */
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 5139a833f721..31b0fabdd2ea 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -181,4 +181,7 @@ int btrfs_balance(struct btrfs_root *dev_root);
 void btrfs_unlock_volumes(void);
 void btrfs_lock_volumes(void);
 int btrfs_chunk_readonly(struct btrfs_root *root, u64 chunk_offset);
+int find_free_dev_extent(struct btrfs_trans_handle *trans,
+                         struct btrfs_device *device, u64 num_bytes,
+                         u64 *start, u64 *max_avail);
 #endif