btrfs: qgroup: Make snapshot accounting work with new extent-oriented
authorQu Wenruo <quwenruo@cn.fujitsu.com>
Mon, 20 Apr 2015 02:09:06 +0000 (10:09 +0800)
committerChris Mason <clm@fb.com>
Wed, 10 Jun 2015 16:26:29 +0000 (09:26 -0700)
qgroup.

Make snapshot accounting work with new extent-oriented mechanism by
skipping given root in new/old_roots in create_pending_snapshot().

Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
fs/btrfs/transaction.c

index 3e3793dcb4c235ee1cab8c1bc29edd2d6750db8d..c0f18e7266b673c22357ebb033e40a4ec6e28b60 100644 (file)
@@ -1295,6 +1295,12 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
        if (pending->error)
                goto no_free_objectid;
 
+       /*
+        * Make qgroup to skip current new snapshot's qgroupid, as it is
+        * accounted by later btrfs_qgroup_inherit().
+        */
+       btrfs_set_skip_qgroup(trans, objectid);
+
        btrfs_reloc_pre_snapshot(trans, pending, &to_reserve);
 
        if (to_reserve > 0) {
@@ -1303,7 +1309,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
                                                     to_reserve,
                                                     BTRFS_RESERVE_NO_FLUSH);
                if (pending->error)
-                       goto no_free_objectid;
+                       goto clear_skip_qgroup;
        }
 
        key.objectid = objectid;
@@ -1401,25 +1407,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
                btrfs_abort_transaction(trans, root, ret);
                goto fail;
        }
-
-       /*
-        * We need to flush delayed refs in order to make sure all of our quota
-        * operations have been done before we call btrfs_qgroup_inherit.
-        */
-       ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
-       if (ret) {
-               btrfs_abort_transaction(trans, root, ret);
-               goto fail;
-       }
-
-       ret = btrfs_qgroup_inherit(trans, fs_info,
-                                  root->root_key.objectid,
-                                  objectid, pending->inherit);
-       if (ret) {
-               btrfs_abort_transaction(trans, root, ret);
-               goto fail;
-       }
-
        /* see comments in should_cow_block() */
        set_bit(BTRFS_ROOT_FORCE_COW, &root->state);
        smp_wmb();
@@ -1502,11 +1489,37 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
                        goto fail;
                }
        }
+
+       ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
+       if (ret) {
+               btrfs_abort_transaction(trans, root, ret);
+               goto fail;
+       }
+
+       /*
+        * account qgroup counters before qgroup_inherit()
+        */
+       ret = btrfs_qgroup_prepare_account_extents(trans, fs_info);
+       if (ret)
+               goto fail;
+       ret = btrfs_qgroup_account_extents(trans, fs_info);
+       if (ret)
+               goto fail;
+       ret = btrfs_qgroup_inherit(trans, fs_info,
+                                  root->root_key.objectid,
+                                  objectid, pending->inherit);
+       if (ret) {
+               btrfs_abort_transaction(trans, root, ret);
+               goto fail;
+       }
+
 fail:
        pending->error = ret;
 dir_item_existed:
        trans->block_rsv = rsv;
        trans->bytes_reserved = 0;
+clear_skip_qgroup:
+       btrfs_clear_skip_qgroup(trans);
 no_free_objectid:
        kfree(new_root_item);
 root_item_alloc_fail: