ocfs2: wait for orphan recovery first once append O_DIRECT write crash

If one node has crashed with orphan entry leftover, another node which do
append O_DIRECT write to the same file will override the
i_dio_orphaned_slot.  Then the old entry won't be cleaned forever.  If
this case happens, we let it wait for orphan recovery first.

Signed-off-by: Joseph Qi <joseph.qi@huawei.com>
Cc: Weiwei Wang <wangww631@huawei.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Xuejiufei <xuejiufei@huawei.com>
Cc: alex chen <alex.chen@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

4 files changed, 43 insertions, 0 deletions

diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h
index ca3431ee7f24..5e86b247c821 100644
--- a/fs/ocfs2/inode.h
+++ b/fs/ocfs2/inode.h
@@ -81,6 +81,8 @@ struct ocfs2_inode_info
 	tid_t i_sync_tid;
 	tid_t i_datasync_tid;
 
+	wait_queue_head_t append_dio_wq;
+
 	struct dquot *i_dquot[MAXQUOTAS];
 };
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 9730f5350ef4..ff531928269e 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -2204,6 +2204,8 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb,
 			ret = ocfs2_del_inode_from_orphan(osb, inode, 0, 0);
 			if (ret)
 				mlog_errno(ret);
+
+			wake_up(&OCFS2_I(inode)->append_dio_wq);
 		} /* else if ORPHAN_NO_NEED_TRUNCATE, do nothing */
 
 next:
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 7eec45d0d85f..b5c3a5ea3ee6 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -2577,6 +2577,27 @@ leave:
 	return status;
 }
 
+static int ocfs2_dio_orphan_recovered(struct inode *inode)
+{
+	int ret;
+	struct buffer_head *di_bh = NULL;
+	struct ocfs2_dinode *di = NULL;
+
+	ret = ocfs2_inode_lock(inode, &di_bh, 1);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return 0;
+	}
+
+	di = (struct ocfs2_dinode *) di_bh->b_data;
+	ret = !(di->i_flags & cpu_to_le32(OCFS2_DIO_ORPHANED_FL));
+	ocfs2_inode_unlock(inode, 1);
+	brelse(di_bh);
+
+	return ret;
+}
+
+#define OCFS2_DIO_ORPHANED_FL_CHECK_INTERVAL 10000
 int ocfs2_add_inode_to_orphan(struct ocfs2_super *osb,
 	struct inode *inode)
 {
@@ -2586,13 +2607,29 @@ int ocfs2_add_inode_to_orphan(struct ocfs2_super *osb,
 	struct buffer_head *di_bh = NULL;
 	int status = 0;
 	handle_t *handle = NULL;
+	struct ocfs2_dinode *di = NULL;
 
+restart:
 	status = ocfs2_inode_lock(inode, &di_bh, 1);
 	if (status < 0) {
 		mlog_errno(status);
 		goto bail;
 	}
 
+	di = (struct ocfs2_dinode *) di_bh->b_data;
+	/*
+	 * Another append dio crashed?
+	 * If so, wait for recovery first.
+	 */
+	if (unlikely(di->i_flags & cpu_to_le32(OCFS2_DIO_ORPHANED_FL))) {
+		ocfs2_inode_unlock(inode, 1);
+		brelse(di_bh);
+		wait_event_interruptible_timeout(OCFS2_I(inode)->append_dio_wq,
+				ocfs2_dio_orphan_recovered(inode),
+				msecs_to_jiffies(OCFS2_DIO_ORPHANED_FL_CHECK_INTERVAL));
+		goto restart;
+	}
+
 	status = ocfs2_prepare_orphan_dir(osb, &orphan_dir_inode,
 			OCFS2_I(inode)->ip_blkno,
 			orphan_name,
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 87a1f7679d9b..26675185b886 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1746,6 +1746,8 @@ static void ocfs2_inode_init_once(void *data)
 	ocfs2_lock_res_init_once(&oi->ip_inode_lockres);
 	ocfs2_lock_res_init_once(&oi->ip_open_lockres);
 
+	init_waitqueue_head(&oi->append_dio_wq);
+
 	ocfs2_metadata_cache_init(INODE_CACHE(&oi->vfs_inode),
 				  &ocfs2_inode_caching_ops);