提交 27eb13e9 编写于 作者: P Philipp Reisner

drbd: Fixed processing of disk-barrier, disk-flushes and disk-drain

Since drbd_bump_write_ordering() is called in the attaching
process while the disk state is D_ATTACHING, it was not
considering these three flags during attach.

A call to this function was missing form drbd_adm_disk_opts().

Fixed both issues.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>
上级 9ed57dcb
......@@ -1229,6 +1229,9 @@ int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info)
}
mutex_unlock(&mdev->tconn->conf_update);
drbd_bump_write_ordering(mdev->tconn, WO_bdev_flush);
drbd_md_sync(mdev);
if (mdev->state.conn >= C_CONNECTED)
......
......@@ -1229,7 +1229,7 @@ void drbd_bump_write_ordering(struct drbd_tconn *tconn, enum write_ordering_e wo
wo = min(pwo, wo);
rcu_read_lock();
idr_for_each_entry(&tconn->volumes, mdev, vnr) {
if (!get_ldev(mdev))
if (!get_ldev_if_state(mdev, D_ATTACHING))
continue;
dc = rcu_dereference(mdev->ldev->disk_conf);
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册