ceph: delete stale dentry when last reference is dropped

introduce ceph_d_delete(), which checks if dentry has valid lease.

Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

fs/ceph/dir.c

@@ -1139,30 +1139,45 @@ void ceph_invalidate_dentry_lease(struct dentry *dentry)
  * Check if dentry lease is valid.  If not, delete the lease.  Try to
  * renew if the least is more than half up.
  */
+static bool __dentry_lease_is_valid(struct ceph_dentry_info *di)
+{
+	struct ceph_mds_session *session;
+
+	if (!di->lease_gen)
+		return false;
+
+	session = di->lease_session;
+	if (session) {
+		u32 gen;
+		unsigned long ttl;
+
+		spin_lock(&session->s_gen_ttl_lock);
+		gen = session->s_cap_gen;
+		ttl = session->s_cap_ttl;
+		spin_unlock(&session->s_gen_ttl_lock);
+
+		if (di->lease_gen == gen &&
+		    time_before(jiffies, ttl) &&
+		    time_before(jiffies, di->time))
+			return true;
+	}
+	di->lease_gen = 0;
+	return false;
+}
+
 static int dentry_lease_is_valid(struct dentry *dentry, unsigned int flags,
 				 struct inode *dir)
 {
 	struct ceph_dentry_info *di;
-	struct ceph_mds_session *s;
-	int valid = 0;
-	u32 gen;
-	unsigned long ttl;
 	struct ceph_mds_session *session = NULL;
 	u32 seq = 0;
+	int valid = 0;
 
 	spin_lock(&dentry->d_lock);
 	di = ceph_dentry(dentry);
-	if (di && di->lease_session) {
-		s = di->lease_session;
-		spin_lock(&s->s_gen_ttl_lock);
-		gen = s->s_cap_gen;
-		ttl = s->s_cap_ttl;
-		spin_unlock(&s->s_gen_ttl_lock);
-
-		if (di->lease_gen == gen &&
-		    time_before(jiffies, di->time) &&
-		    time_before(jiffies, ttl)) {
+	if (di && __dentry_lease_is_valid(di)) {
 		valid = 1;
+
 		if (di->lease_renew_after &&
 		    time_after(jiffies, di->lease_renew_after)) {
 			/*
@@ -1173,14 +1188,13 @@ static int dentry_lease_is_valid(struct dentry *dentry, unsigned int flags,
 			if (flags & LOOKUP_RCU) {
 				valid = -ECHILD;
 			} else {
-					session = ceph_get_mds_session(s);
+				session = ceph_get_mds_session(di->lease_session);
 				seq = di->lease_seq;
 				di->lease_renew_after = 0;
 				di->lease_renew_from = jiffies;
 			}
 		}
 	}
-	}
 	spin_unlock(&dentry->d_lock);
 
 	if (session) {
@@ -1192,6 +1206,38 @@ static int dentry_lease_is_valid(struct dentry *dentry, unsigned int flags,
 	return valid;
 }
 
+/*
+ * Called under dentry->d_lock.
+ */
+static int __dir_lease_try_check(const struct dentry *dentry)
+{
+	struct ceph_dentry_info *di = ceph_dentry(dentry);
+	struct inode *dir;
+	struct ceph_inode_info *ci;
+	int valid = 0;
+
+	if (!di->lease_shared_gen)
+		return 0;
+	if (IS_ROOT(dentry))
+		return 0;
+
+	dir = d_inode(dentry->d_parent);
+	ci = ceph_inode(dir);
+
+	if (spin_trylock(&ci->i_ceph_lock)) {
+		if (atomic_read(&ci->i_shared_gen) == di->lease_shared_gen &&
+		    __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 0))
+			valid = 1;
+		spin_unlock(&ci->i_ceph_lock);
+	} else {
+		valid = -EBUSY;
+	}
+
+	if (!valid)
+		di->lease_shared_gen = 0;
+	return valid;
+}
+
 /*
  * Check if directory-wide content lease/cap is valid.
  */
@@ -1308,6 +1354,31 @@ static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags)
 	return valid;
 }
 
+/*
+ * Delete unused dentry that doesn't have valid lease
+ *
+ * Called under dentry->d_lock.
+ */
+static int ceph_d_delete(const struct dentry *dentry)
+{
+	struct ceph_dentry_info *di;
+
+	/* won't release caps */
+	if (d_really_is_negative(dentry))
+		return 0;
+	if (ceph_snap(d_inode(dentry)) != CEPH_NOSNAP)
+		return 0;
+	/* vaild lease? */
+	di = ceph_dentry(dentry);
+	if (di) {
+		if (__dentry_lease_is_valid(di))
+			return 0;
+		if (__dir_lease_try_check(dentry))
+			return 0;
+	}
+	return 1;
+}
+
 /*
  * Release our ceph_dentry_info.
  */
@@ -1531,6 +1602,7 @@ const struct inode_operations ceph_snapdir_iops = {
 
 const struct dentry_operations ceph_dentry_ops = {
 	.d_revalidate = ceph_d_revalidate,
+	.d_delete = ceph_d_delete,
 	.d_release = ceph_d_release,
 	.d_prune = ceph_d_prune,
 	.d_init = ceph_d_init,

fs/ceph/inode.c

@@ -497,7 +497,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_wrbuffer_ref = 0;
 	ci->i_wrbuffer_ref_head = 0;
 	atomic_set(&ci->i_filelock_ref, 0);
-	atomic_set(&ci->i_shared_gen, 0);
+	atomic_set(&ci->i_shared_gen, 1);
 	ci->i_rdcache_gen = 0;
 	ci->i_rdcache_revoking = 0;
 

fs/ceph/mds_client.c

@@ -621,7 +621,7 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
 	ceph_con_init(&s->s_con, s, &mds_con_ops, &mdsc->fsc->client->msgr);
 
 	spin_lock_init(&s->s_gen_ttl_lock);
-	s->s_cap_gen = 0;
+	s->s_cap_gen = 1;
 	s->s_cap_ttl = jiffies - 1;
 
 	spin_lock_init(&s->s_cap_lock);

fs/ceph/super.h

@@ -594,7 +594,7 @@ extern u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
 			    struct ceph_inode_frag *pfrag,
 			    int *found);
 
-static inline struct ceph_dentry_info *ceph_dentry(struct dentry *dentry)
+static inline struct ceph_dentry_info *ceph_dentry(const struct dentry *dentry)
 {
 	return (struct ceph_dentry_info *)dentry->d_fsdata;
 }