Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel…

…/git/sage/ceph-client

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  ceph: use separate class for ceph sockets' sk_lock
  ceph: reserve one more caps space when doing readdir
  ceph: queue_cap_snap should always queue dirty context
  ceph: fix dentry reference leak in dcache readdir
  ceph: decode v5 of osdmap (pool names) [protocol change]
  ceph: fix ack counter reset on connection reset
  ceph: fix leaked inode ref due to snap metadata writeback race
  ceph: fix snap context reference leaks
  ceph: allow writeback of snapped pages older than 'oldest' snapc
  ceph: fix dentry rehashing on virtual .snap dir

fs/ceph/addr.c

@@ -337,16 +337,15 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
 /*
  * Get ref for the oldest snapc for an inode with dirty data... that is, the
  * only snap context we are allowed to write back.
- *
- * Caller holds i_lock.
  */
-static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
-						      u64 *snap_size)
+static struct ceph_snap_context *get_oldest_context(struct inode *inode,
+						    u64 *snap_size)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_snap_context *snapc = NULL;
 	struct ceph_cap_snap *capsnap = NULL;
 
+	spin_lock(&inode->i_lock);
 	list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
 		dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap,
 		     capsnap->context, capsnap->dirty_pages);
@@ -357,21 +356,11 @@ static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
 			break;
 		}
 	}
-	if (!snapc && ci->i_snap_realm) {
-		snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);
+	if (!snapc && ci->i_head_snapc) {
+		snapc = ceph_get_snap_context(ci->i_head_snapc);
 		dout(" head snapc %p has %d dirty pages\n",
 		     snapc, ci->i_wrbuffer_ref_head);
 	}
-	return snapc;
-}
-
-static struct ceph_snap_context *get_oldest_context(struct inode *inode,
-						    u64 *snap_size)
-{
-	struct ceph_snap_context *snapc = NULL;
-
-	spin_lock(&inode->i_lock);
-	snapc = __get_oldest_context(inode, snap_size);
 	spin_unlock(&inode->i_lock);
 	return snapc;
 }
@@ -392,7 +381,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	int len = PAGE_CACHE_SIZE;
 	loff_t i_size;
 	int err = 0;
-	struct ceph_snap_context *snapc;
+	struct ceph_snap_context *snapc, *oldest;
 	u64 snap_size = 0;
 	long writeback_stat;
 
@@ -413,13 +402,16 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 		dout("writepage %p page %p not dirty?\n", inode, page);
 		goto out;
 	}
-	if (snapc != get_oldest_context(inode, &snap_size)) {
+	oldest = get_oldest_context(inode, &snap_size);
+	if (snapc->seq > oldest->seq) {
 		dout("writepage %p page %p snapc %p not writeable - noop\n",
 		     inode, page, (void *)page->private);
 		/* we should only noop if called by kswapd */
 		WARN_ON((current->flags & PF_MEMALLOC) == 0);
+		ceph_put_snap_context(oldest);
 		goto out;
 	}
+	ceph_put_snap_context(oldest);
 
 	/* is this a partial page at end of file? */
 	if (snap_size)
@@ -458,7 +450,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	ClearPagePrivate(page);
 	end_page_writeback(page);
 	ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
-	ceph_put_snap_context(snapc);
+	ceph_put_snap_context(snapc);  /* page's reference */
 out:
 	return err;
 }
@@ -558,9 +550,9 @@ static void writepages_finish(struct ceph_osd_request *req,
 			dout("inode %p skipping page %p\n", inode, page);
 			wbc->pages_skipped++;
 		}
+		ceph_put_snap_context((void *)page->private);
 		page->private = 0;
 		ClearPagePrivate(page);
-		ceph_put_snap_context(snapc);
 		dout("unlocking %d %p\n", i, page);
 		end_page_writeback(page);
 
@@ -618,7 +610,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 	int range_whole = 0;
 	int should_loop = 1;
 	pgoff_t max_pages = 0, max_pages_ever = 0;
-	struct ceph_snap_context *snapc = NULL, *last_snapc = NULL;
+	struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc;
 	struct pagevec pvec;
 	int done = 0;
 	int rc = 0;
@@ -770,9 +762,10 @@ static int ceph_writepages_start(struct address_space *mapping,
 			}
 
 			/* only if matching snap context */
-			if (snapc != (void *)page->private) {
-				dout("page snapc %p != oldest %p\n",
-				     (void *)page->private, snapc);
+			pgsnapc = (void *)page->private;
+			if (pgsnapc->seq > snapc->seq) {
+				dout("page snapc %p %lld > oldest %p %lld\n",
+				     pgsnapc, pgsnapc->seq, snapc, snapc->seq);
 				unlock_page(page);
 				if (!locked_pages)
 					continue; /* keep looking for snap */
@@ -914,7 +907,10 @@ static int context_is_writeable_or_written(struct inode *inode,
 					   struct ceph_snap_context *snapc)
 {
 	struct ceph_snap_context *oldest = get_oldest_context(inode, NULL);
-	return !oldest || snapc->seq <= oldest->seq;
+	int ret = !oldest || snapc->seq <= oldest->seq;
+
+	ceph_put_snap_context(oldest);
+	return ret;
 }
 
 /*
@@ -936,8 +932,8 @@ static int ceph_update_writeable_page(struct file *file,
 	int pos_in_page = pos & ~PAGE_CACHE_MASK;
 	int end_in_page = pos_in_page + len;
 	loff_t i_size;
-	struct ceph_snap_context *snapc;
 	int r;
+	struct ceph_snap_context *snapc, *oldest;
 
 retry_locked:
 	/* writepages currently holds page lock, but if we change that later, */
@@ -947,23 +943,24 @@ static int ceph_update_writeable_page(struct file *file,
 	BUG_ON(!ci->i_snap_realm);
 	down_read(&mdsc->snap_rwsem);
 	BUG_ON(!ci->i_snap_realm->cached_context);
-	if (page->private &&
-	    (void *)page->private != ci->i_snap_realm->cached_context) {
+	snapc = (void *)page->private;
+	if (snapc && snapc != ci->i_head_snapc) {
 		/*
 		 * this page is already dirty in another (older) snap
 		 * context!  is it writeable now?
 		 */
-		snapc = get_oldest_context(inode, NULL);
+		oldest = get_oldest_context(inode, NULL);
 		up_read(&mdsc->snap_rwsem);
 
-		if (snapc != (void *)page->private) {
+		if (snapc->seq > oldest->seq) {
+			ceph_put_snap_context(oldest);
 			dout(" page %p snapc %p not current or oldest\n",
-			     page, (void *)page->private);
+			     page, snapc);
 			/*
 			 * queue for writeback, and wait for snapc to
 			 * be writeable or written
 			 */
-			snapc = ceph_get_snap_context((void *)page->private);
+			snapc = ceph_get_snap_context(snapc);
 			unlock_page(page);
 			ceph_queue_writeback(inode);
 			r = wait_event_interruptible(ci->i_cap_wq,
@@ -973,6 +970,7 @@ static int ceph_update_writeable_page(struct file *file,
 				return r;
 			return -EAGAIN;
 		}
+		ceph_put_snap_context(oldest);
 
 		/* yay, writeable, do it now (without dropping page lock) */
 		dout(" page %p snapc %p not current, but oldest\n",

fs/ceph/caps.c

@@ -1205,6 +1205,12 @@ void __ceph_flush_snaps(struct ceph_inode_info *ci,
 		if (capsnap->dirty_pages || capsnap->writing)
 			continue;
 
+		/*
+		 * if cap writeback already occurred, we should have dropped
+		 * the capsnap in ceph_put_wrbuffer_cap_refs.
+		 */
+		BUG_ON(capsnap->dirty == 0);
+
 		/* pick mds, take s_mutex */
 		mds = __ceph_get_cap_mds(ci, &mseq);
 		if (session && session->s_mds != mds) {
@@ -2118,8 +2124,8 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
 		}
 	spin_unlock(&inode->i_lock);
 
-	dout("put_cap_refs %p had %s %s\n", inode, ceph_cap_string(had),
-	     last ? "last" : "");
+	dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had),
+	     last ? " last" : "", put ? " put" : "");
 
 	if (last && !flushsnaps)
 		ceph_check_caps(ci, 0, NULL);
@@ -2143,7 +2149,8 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 {
 	struct inode *inode = &ci->vfs_inode;
 	int last = 0;
-	int last_snap = 0;
+	int complete_capsnap = 0;
+	int drop_capsnap = 0;
 	int found = 0;
 	struct ceph_cap_snap *capsnap = NULL;
 
@@ -2166,30 +2173,45 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
 		list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
 			if (capsnap->context == snapc) {
 				found = 1;
-				capsnap->dirty_pages -= nr;
-				last_snap = !capsnap->dirty_pages;
 				break;
 			}
 		}
 		BUG_ON(!found);
+		capsnap->dirty_pages -= nr;
+		if (capsnap->dirty_pages == 0) {
+			complete_capsnap = 1;
+			if (capsnap->dirty == 0)
+				/* cap writeback completed before we created
+				 * the cap_snap; no FLUSHSNAP is needed */
+				drop_capsnap = 1;
+		}
 		dout("put_wrbuffer_cap_refs on %p cap_snap %p "
-		     " snap %lld %d/%d -> %d/%d %s%s\n",
+		     " snap %lld %d/%d -> %d/%d %s%s%s\n",
 		     inode, capsnap, capsnap->context->seq,
 		     ci->i_wrbuffer_ref+nr, capsnap->dirty_pages + nr,
 		     ci->i_wrbuffer_ref, capsnap->dirty_pages,
 		     last ? " (wrbuffer last)" : "",
-		     last_snap ? " (capsnap last)" : "");
+		     complete_capsnap ? " (complete capsnap)" : "",
+		     drop_capsnap ? " (drop capsnap)" : "");
+		if (drop_capsnap) {
+			ceph_put_snap_context(capsnap->context);
+			list_del(&capsnap->ci_item);
+			list_del(&capsnap->flushing_item);
+			ceph_put_cap_snap(capsnap);
+		}
 	}
 
 	spin_unlock(&inode->i_lock);
 
 	if (last) {
 		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
 		iput(inode);
-	} else if (last_snap) {
+	} else if (complete_capsnap) {
 		ceph_flush_snaps(ci);
 		wake_up(&ci->i_cap_wq);
 	}
+	if (drop_capsnap)
+		iput(inode);
 }
 
 /*
@@ -2465,8 +2487,8 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid,
 				break;
 			}
 			WARN_ON(capsnap->dirty_pages || capsnap->writing);
-			dout(" removing cap_snap %p follows %lld\n",
-			     capsnap, follows);
+			dout(" removing %p cap_snap %p follows %lld\n",
+			     inode, capsnap, follows);
 			ceph_put_snap_context(capsnap->context);
 			list_del(&capsnap->ci_item);
 			list_del(&capsnap->flushing_item);

fs/ceph/dir.c

@@ -171,11 +171,11 @@ static int __dcache_readdir(struct file *filp,
 	spin_lock(&inode->i_lock);
 	spin_lock(&dcache_lock);
 
+	last = dentry;
+
 	if (err < 0)
 		goto out_unlock;
 
-	last = dentry;
-
 	p = p->prev;
 	filp->f_pos++;
 
@@ -312,7 +312,7 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		req->r_readdir_offset = fi->next_offset;
 		req->r_args.readdir.frag = cpu_to_le32(frag);
 		req->r_args.readdir.max_entries = cpu_to_le32(max_entries);
-		req->r_num_caps = max_entries;
+		req->r_num_caps = max_entries + 1;
 		err = ceph_mdsc_do_request(mdsc, NULL, req);
 		if (err < 0) {
 			ceph_mdsc_put_request(req);
@@ -489,6 +489,7 @@ struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
 		struct inode *inode = ceph_get_snapdir(parent);
 		dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n",
 		     dentry, dentry->d_name.len, dentry->d_name.name, inode);
+		BUG_ON(!d_unhashed(dentry));
 		d_add(dentry, inode);
 		err = 0;
 	}

fs/ceph/inode.c

@@ -886,6 +886,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	struct inode *in = NULL;
 	struct ceph_mds_reply_inode *ininfo;
 	struct ceph_vino vino;
+	struct ceph_client *client = ceph_sb_to_client(sb);
 	int i = 0;
 	int err = 0;
 
@@ -949,7 +950,14 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 			return err;
 	}
 
-	if (rinfo->head->is_dentry && !req->r_aborted) {
+	/*
+	 * ignore null lease/binding on snapdir ENOENT, or else we
+	 * will have trouble splicing in the virtual snapdir later
+	 */
+	if (rinfo->head->is_dentry && !req->r_aborted &&
+	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
+					       client->mount_args->snapdir_name,
+					       req->r_dentry->d_name.len))) {
 		/*
 		 * lookup link rename   : null -> possibly existing inode
 		 * mknod symlink mkdir  : null -> new inode

fs/ceph/messenger.c

@@ -30,6 +30,10 @@ static char tag_msg = CEPH_MSGR_TAG_MSG;
 static char tag_ack = CEPH_MSGR_TAG_ACK;
 static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
 
+#ifdef CONFIG_LOCKDEP
+static struct lock_class_key socket_class;
+#endif
+
 
 static void queue_con(struct ceph_connection *con);
 static void con_work(struct work_struct *);
@@ -228,6 +232,10 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con)
 	con->sock = sock;
 	sock->sk->sk_allocation = GFP_NOFS;
 
+#ifdef CONFIG_LOCKDEP
+	lockdep_set_class(&sock->sk->sk_lock, &socket_class);
+#endif
+
 	set_sock_callbacks(sock, con);
 
 	dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
@@ -333,6 +341,7 @@ static void reset_connection(struct ceph_connection *con)
 		con->out_msg = NULL;
 	}
 	con->in_seq = 0;
+	con->in_seq_acked = 0;
 }
 
 /*