Ocfs2: Handle deletion of reflinked oprhan inodes correctly.
The rule is that all inodes in the orphan dir have ORPHANED_FL,
otherwise we treated it as an ERROR. This rule works well except
for some rare cases of reflink operation:
http://oss.oracle.com/bugzilla/show_bug.cgi?id=1215
The problem is caused by how reflink and our orphan_scan thread
interact.
* The orphan scan pulls the orphans into a queue first, then runs the
queue at a later time. We only hold the orphan_dir's lock
during scanning.
* Reflink create a oprhaned target in orphan_dir as its first step.
It removes the target and clears the flag as the final step.
These two steps take the orphan_dir's lock, but it is not held for
the duration.
Based on the above semantics, a reflink inode can be moved out of the
orphan dir and have its ORPHANED_FL cleared before the queue of orphans
is run. This leads to a ERROR in ocfs2_query_wipde_inode().
This patch teaches ocfs2_query_wipe_inode() to detect previously
orphaned reflink targets. If a reflink fails or a crash occurs during
the relfink operation, the inode will retain ORPHANED_FL and will be
properly wiped.
Signed-off-by: Tristan Ye <tristan.ye@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 278a223..ab20790 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -891,6 +891,21 @@
/* Do some basic inode verification... */
di = (struct ocfs2_dinode *) di_bh->b_data;
if (!(di->i_flags & cpu_to_le32(OCFS2_ORPHANED_FL))) {
+ /*
+ * Inodes in the orphan dir must have ORPHANED_FL. The only
+ * inodes that come back out of the orphan dir are reflink
+ * targets. A reflink target may be moved out of the orphan
+ * dir between the time we scan the directory and the time we
+ * process it. This would lead to HAS_REFCOUNT_FL being set but
+ * ORPHANED_FL not.
+ */
+ if (di->i_dyn_features & cpu_to_le16(OCFS2_HAS_REFCOUNT_FL)) {
+ mlog(0, "Reflinked inode %llu is no longer orphaned. "
+ "it shouldn't be deleted\n",
+ (unsigned long long)oi->ip_blkno);
+ goto bail;
+ }
+
/* for lack of a better error? */
status = -EEXIST;
mlog(ML_ERROR,