Merge tag 'dlm-5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm

Pull dlm updates from David Teigland:
 "This includes more dlm networking cleanups and improvements for making
  dlm shutdowns more robust"

* tag 'dlm-5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm:
  fs: dlm: fix missing unlock on error in accept_from_sock()
  fs: dlm: add shutdown hook
  fs: dlm: flush swork on shutdown
  fs: dlm: remove unaligned memory access handling
  fs: dlm: check on minimum msglen size
  fs: dlm: simplify writequeue handling
  fs: dlm: use GFP_ZERO for page buffer
  fs: dlm: change allocation limits
  fs: dlm: add check if dlm is currently running
  fs: dlm: add errno handling to check callback
  fs: dlm: set subclass for othercon sock_mutex
  fs: dlm: set connected bit after accept
  fs: dlm: fix mark setting deadlock
  fs: dlm: fix debugfs dump

9 files changed, 202 insertions, 142 deletions

diff --git a/fs/dlm/config.c b/fs/dlm/config.c
index 49c5f9407098..88d95d96e36c 100644
--- a/fs/dlm/config.c
+++ b/fs/dlm/config.c
@@ -125,7 +125,7 @@ static ssize_t cluster_cluster_name_store(struct config_item *item,
 CONFIGFS_ATTR(cluster_, cluster_name);
 
 static ssize_t cluster_set(struct dlm_cluster *cl, unsigned int *cl_field,
-			   int *info_field, bool (*check_cb)(unsigned int x),
+			   int *info_field, int (*check_cb)(unsigned int x),
 			   const char *buf, size_t len)
 {
 	unsigned int x;
@@ -137,8 +137,11 @@ static ssize_t cluster_set(struct dlm_cluster *cl, unsigned int *cl_field,
 	if (rc)
 		return rc;
 
-	if (check_cb && check_cb(x))
-		return -EINVAL;
+	if (check_cb) {
+		rc = check_cb(x);
+		if (rc)
+			return rc;
+	}
 
 	*cl_field = x;
 	*info_field = x;
@@ -161,17 +164,53 @@ static ssize_t cluster_##name##_show(struct config_item *item, char *buf)     \
 }                                                                             \
 CONFIGFS_ATTR(cluster_, name);
 
-static bool dlm_check_zero(unsigned int x)
+static int dlm_check_protocol_and_dlm_running(unsigned int x)
+{
+	switch (x) {
+	case 0:
+		/* TCP */
+		break;
+	case 1:
+		/* SCTP */
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (dlm_allow_conn)
+		return -EBUSY;
+
+	return 0;
+}
+
+static int dlm_check_zero_and_dlm_running(unsigned int x)
+{
+	if (!x)
+		return -EINVAL;
+
+	if (dlm_allow_conn)
+		return -EBUSY;
+
+	return 0;
+}
+
+static int dlm_check_zero(unsigned int x)
 {
-	return !x;
+	if (!x)
+		return -EINVAL;
+
+	return 0;
 }
 
-static bool dlm_check_buffer_size(unsigned int x)
+static int dlm_check_buffer_size(unsigned int x)
 {
-	return (x < DEFAULT_BUFFER_SIZE);
+	if (x < DEFAULT_BUFFER_SIZE)
+		return -EINVAL;
+
+	return 0;
 }
 
-CLUSTER_ATTR(tcp_port, dlm_check_zero);
+CLUSTER_ATTR(tcp_port, dlm_check_zero_and_dlm_running);
 CLUSTER_ATTR(buffer_size, dlm_check_buffer_size);
 CLUSTER_ATTR(rsbtbl_size, dlm_check_zero);
 CLUSTER_ATTR(recover_timer, dlm_check_zero);
@@ -179,7 +218,7 @@ CLUSTER_ATTR(toss_secs, dlm_check_zero);
 CLUSTER_ATTR(scan_secs, dlm_check_zero);
 CLUSTER_ATTR(log_debug, NULL);
 CLUSTER_ATTR(log_info, NULL);
-CLUSTER_ATTR(protocol, NULL);
+CLUSTER_ATTR(protocol, dlm_check_protocol_and_dlm_running);
 CLUSTER_ATTR(mark, NULL);
 CLUSTER_ATTR(timewarn_cs, dlm_check_zero);
 CLUSTER_ATTR(waitwarn_us, NULL);
@@ -688,6 +727,7 @@ static ssize_t comm_mark_show(struct config_item *item, char *buf)
 static ssize_t comm_mark_store(struct config_item *item, const char *buf,
 			       size_t len)
 {
+	struct dlm_comm *comm;
 	unsigned int mark;
 	int rc;
 
@@ -695,7 +735,15 @@ static ssize_t comm_mark_store(struct config_item *item, const char *buf,
 	if (rc)
 		return rc;
 
-	config_item_to_comm(item)->mark = mark;
+	if (mark == 0)
+		mark = dlm_config.ci_mark;
+
+	comm = config_item_to_comm(item);
+	rc = dlm_lowcomms_nodes_set_mark(comm->nodeid, mark);
+	if (rc)
+		return rc;
+
+	comm->mark = mark;
 	return len;
 }
 
@@ -870,24 +918,6 @@ int dlm_comm_seq(int nodeid, uint32_t *seq)
 	return 0;
 }
 
-void dlm_comm_mark(int nodeid, unsigned int *mark)
-{
-	struct dlm_comm *cm;
-
-	cm = get_comm(nodeid);
-	if (!cm) {
-		*mark = dlm_config.ci_mark;
-		return;
-	}
-
-	if (cm->mark)
-		*mark = cm->mark;
-	else
-		*mark = dlm_config.ci_mark;
-
-	put_comm(cm);
-}
-
 int dlm_our_nodeid(void)
 {
 	return local_comm ? local_comm->nodeid : 0;
diff --git a/fs/dlm/config.h b/fs/dlm/config.h
index c210250a2581..d2cd4bd20313 100644
--- a/fs/dlm/config.h
+++ b/fs/dlm/config.h
@@ -48,7 +48,6 @@ void dlm_config_exit(void);
 int dlm_config_nodes(char *lsname, struct dlm_config_node **nodes_out,
 		     int *count_out);
 int dlm_comm_seq(int nodeid, uint32_t *seq);
-void dlm_comm_mark(int nodeid, unsigned int *mark);
 int dlm_our_nodeid(void);
 int dlm_our_addr(struct sockaddr_storage *addr, int num);
 
diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c
index d6bbccb0ed15..d5bd990bcab8 100644
--- a/fs/dlm/debug_fs.c
+++ b/fs/dlm/debug_fs.c
@@ -542,6 +542,7 @@ static void *table_seq_next(struct seq_file *seq, void *iter_ptr, loff_t *pos)
 
 		if (bucket >= ls->ls_rsbtbl_size) {
 			kfree(ri);
+			++*pos;
 			return NULL;
 		}
 		tree = toss ? &ls->ls_rsbtbl[bucket].toss : &ls->ls_rsbtbl[bucket].keep;
diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c
index 002123efc6b0..b93df39d0915 100644
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -3541,8 +3541,6 @@ static int _create_message(struct dlm_ls *ls, int mb_len,
 	if (!mh)
 		return -ENOBUFS;
 
-	memset(mb, 0, mb_len);
-
 	ms = (struct dlm_message *) mb;
 
 	ms->m_header.h_version = (DLM_HEADER_MAJOR | DLM_HEADER_MINOR);
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index 561dcad08ad6..c14cf2b7faab 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -404,12 +404,6 @@ static int threads_start(void)
 	return error;
 }
 
-static void threads_stop(void)
-{
-	dlm_scand_stop();
-	dlm_lowcomms_stop();
-}
-
 static int new_lockspace(const char *name, const char *cluster,
 			 uint32_t flags, int lvblen,
 			 const struct dlm_lockspace_ops *ops, void *ops_arg,
@@ -702,8 +696,11 @@ int dlm_new_lockspace(const char *name, const char *cluster,
 		ls_count++;
 	if (error > 0)
 		error = 0;
-	if (!ls_count)
-		threads_stop();
+	if (!ls_count) {
+		dlm_scand_stop();
+		dlm_lowcomms_shutdown();
+		dlm_lowcomms_stop();
+	}
  out:
 	mutex_unlock(&ls_lock);
 	return error;
@@ -788,6 +785,11 @@ static int release_lockspace(struct dlm_ls *ls, int force)
 
 	dlm_recoverd_stop(ls);
 
+	if (ls_count == 1) {
+		dlm_scand_stop();
+		dlm_lowcomms_shutdown();
+	}
+
 	dlm_callback_stop(ls);
 
 	remove_lockspace(ls);
@@ -880,7 +882,7 @@ int dlm_release_lockspace(void *lockspace, int force)
 	if (!error)
 		ls_count--;
 	if (!ls_count)
-		threads_stop();
+		dlm_lowcomms_stop();
 	mutex_unlock(&ls_lock);
 
 	return error;
diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index 372c34ff8594..166e36fcf3e4 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -102,6 +102,9 @@ struct listen_connection {
 	struct work_struct rwork;
 };
 
+#define DLM_WQ_REMAIN_BYTES(e) (PAGE_SIZE - e->end)
+#define DLM_WQ_LENGTH_BYTES(e) (e->end - e->offset)
+
 /* An entry waiting to be sent */
 struct writequeue_entry {
 	struct list_head list;
@@ -116,6 +119,7 @@ struct writequeue_entry {
 struct dlm_node_addr {
 	struct list_head list;
 	int nodeid;
+	int mark;
 	int addr_count;
 	int curr_addr_index;
 	struct sockaddr_storage *addr[DLM_MAX_ADDR_COUNT];
@@ -134,7 +138,7 @@ static DEFINE_SPINLOCK(dlm_node_addrs_spin);
 static struct listen_connection listen_con;
 static struct sockaddr_storage *dlm_local_addr[DLM_MAX_ADDR_COUNT];
 static int dlm_local_count;
-static int dlm_allow_conn;
+int dlm_allow_conn;
 
 /* Work queues */
 static struct workqueue_struct *recv_workqueue;
@@ -303,7 +307,8 @@ static int addr_compare(const struct sockaddr_storage *x,
 }
 
 static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out,
-			  struct sockaddr *sa_out, bool try_new_addr)
+			  struct sockaddr *sa_out, bool try_new_addr,
+			  unsigned int *mark)
 {
 	struct sockaddr_storage sas;
 	struct dlm_node_addr *na;
@@ -331,6 +336,8 @@ static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out,
 	if (!na->addr_count)
 		return -ENOENT;
 
+	*mark = na->mark;
+
 	if (sas_out)
 		memcpy(sas_out, &sas, sizeof(struct sockaddr_storage));
 
@@ -350,7 +357,8 @@ static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out,
 	return 0;
 }
 
-static int addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid)
+static int addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid,
+			  unsigned int *mark)
 {
 	struct dlm_node_addr *na;
 	int rv = -EEXIST;
@@ -364,6 +372,7 @@ static int addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid)
 		for (addr_i = 0; addr_i < na->addr_count; addr_i++) {
 			if (addr_compare(na->addr[addr_i], addr)) {
 				*nodeid = na->nodeid;
+				*mark = na->mark;
 				rv = 0;
 				goto unlock;
 			}
@@ -412,6 +421,7 @@ int dlm_lowcomms_addr(int nodeid, struct sockaddr_storage *addr, int len)
 		new_node->nodeid = nodeid;
 		new_node->addr[0] = new_addr;
 		new_node->addr_count = 1;
+		new_node->mark = dlm_config.ci_mark;
 		list_add(&new_node->list, &dlm_node_addrs);
 		spin_unlock(&dlm_node_addrs_spin);
 		return 0;
@@ -519,6 +529,23 @@ int dlm_lowcomms_connect_node(int nodeid)
 	return 0;
 }
 
+int dlm_lowcomms_nodes_set_mark(int nodeid, unsigned int mark)
+{
+	struct dlm_node_addr *na;
+
+	spin_lock(&dlm_node_addrs_spin);
+	na = find_node_addr(nodeid);
+	if (!na) {
+		spin_unlock(&dlm_node_addrs_spin);
+		return -ENOENT;
+	}
+
+	na->mark = mark;
+	spin_unlock(&dlm_node_addrs_spin);
+
+	return 0;
+}
+
 static void lowcomms_error_report(struct sock *sk)
 {
 	struct connection *con;
@@ -685,10 +712,7 @@ static void shutdown_connection(struct connection *con)
 {
 	int ret;
 
-	if (cancel_work_sync(&con->swork)) {
-		log_print("canceled swork for node %d", con->nodeid);
-		clear_bit(CF_WRITE_PENDING, &con->flags);
-	}
+	flush_work(&con->swork);
 
 	mutex_lock(&con->sock_mutex);
 	/* nothing to shutdown */
@@ -867,7 +891,7 @@ static int accept_from_sock(struct listen_connection *con)
 
 	/* Get the new node's NODEID */
 	make_sockaddr(&peeraddr, 0, &len);
-	if (addr_to_nodeid(&peeraddr, &nodeid)) {
+	if (addr_to_nodeid(&peeraddr, &nodeid, &mark)) {
 		unsigned char *b=(unsigned char *)&peeraddr;
 		log_print("connect from non cluster node");
 		print_hex_dump_bytes("ss: ", DUMP_PREFIX_NONE, 
@@ -876,9 +900,6 @@ static int accept_from_sock(struct listen_connection *con)
 		return -1;
 	}
 
-	dlm_comm_mark(nodeid, &mark);
-	sock_set_mark(newsock->sk, mark);
-
 	log_print("got connection from %d", nodeid);
 
 	/*  Check to see if we already have a connection to this node. This
@@ -892,6 +913,8 @@ static int accept_from_sock(struct listen_connection *con)
 		goto accept_err;
 	}
 
+	sock_set_mark(newsock->sk, mark);
+
 	mutex_lock(&newcon->sock_mutex);
 	if (newcon->sock) {
 		struct connection *othercon = newcon->othercon;
@@ -908,16 +931,18 @@ static int accept_from_sock(struct listen_connection *con)
 			result = dlm_con_init(othercon, nodeid);
 			if (result < 0) {
 				kfree(othercon);
+				mutex_unlock(&newcon->sock_mutex);
 				goto accept_err;
 			}
 
+			lockdep_set_subclass(&othercon->sock_mutex, 1);
 			newcon->othercon = othercon;
 		} else {
 			/* close other sock con if we have something new */
 			close_connection(othercon, false, true, false);
 		}
 
-		mutex_lock_nested(&othercon->sock_mutex, 1);
+		mutex_lock(&othercon->sock_mutex);
 		add_sock(newsock, othercon);
 		addcon = othercon;
 		mutex_unlock(&othercon->sock_mutex);
@@ -930,6 +955,7 @@ static int accept_from_sock(struct listen_connection *con)
 		addcon = newcon;
 	}
 
+	set_bit(CF_CONNECTED, &addcon->flags);
 	mutex_unlock(&newcon->sock_mutex);
 
 	/*
@@ -1015,8 +1041,6 @@ static void sctp_connect_to_sock(struct connection *con)
 	struct socket *sock;
 	unsigned int mark;
 
-	dlm_comm_mark(con->nodeid, &mark);
-
 	mutex_lock(&con->sock_mutex);
 
 	/* Some odd races can cause double-connects, ignore them */
@@ -1029,7 +1053,7 @@ static void sctp_connect_to_sock(struct connection *con)
 	}
 
 	memset(&daddr, 0, sizeof(daddr));
-	result = nodeid_to_addr(con->nodeid, &daddr, NULL, true);
+	result = nodeid_to_addr(con->nodeid, &daddr, NULL, true, &mark);
 	if (result < 0) {
 		log_print("no address for nodeid %d", con->nodeid);
 		goto out;
@@ -1104,13 +1128,11 @@ out:
 static void tcp_connect_to_sock(struct connection *con)
 {
 	struct sockaddr_storage saddr, src_addr;
+	unsigned int mark;
 	int addr_len;
 	struct socket *sock = NULL;
-	unsigned int mark;
 	int result;
 
-	dlm_comm_mark(con->nodeid, &mark);
-
 	mutex_lock(&con->sock_mutex);
 	if (con->retries++ > MAX_CONNECT_RETRIES)
 		goto out;
@@ -1125,15 +1147,15 @@ static void tcp_connect_to_sock(struct connection *con)
 	if (result < 0)
 		goto out_err;
 
-	sock_set_mark(sock->sk, mark);
-
 	memset(&saddr, 0, sizeof(saddr));
-	result = nodeid_to_addr(con->nodeid, &saddr, NULL, false);
+	result = nodeid_to_addr(con->nodeid, &saddr, NULL, false, &mark);
 	if (result < 0) {
 		log_print("no address for nodeid %d", con->nodeid);
 		goto out_err;
 	}
 
+	sock_set_mark(sock->sk, mark);
+
 	add_sock(sock, con);
 
 	/* Bind to our cluster-known address connecting to avoid
@@ -1330,70 +1352,72 @@ static struct writequeue_entry *new_writequeue_entry(struct connection *con,
 {
 	struct writequeue_entry *entry;
 
-	entry = kmalloc(sizeof(struct writequeue_entry), allocation);
+	entry = kzalloc(sizeof(*entry), allocation);
 	if (!entry)
 		return NULL;
 
-	entry->page = alloc_page(allocation);
+	entry->page = alloc_page(allocation | __GFP_ZERO);
 	if (!entry->page) {
 		kfree(entry);
 		return NULL;
 	}
 
-	entry->offset = 0;
-	entry->len = 0;
-	entry->end = 0;
-	entry->users = 0;
 	entry->con = con;
+	entry->users = 1;
 
 	return entry;
 }
 
-void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc)
+static struct writequeue_entry *new_wq_entry(struct connection *con, int len,
+					     gfp_t allocation, char **ppc)
 {
-	struct connection *con;
 	struct writequeue_entry *e;
-	int offset = 0;
 
-	if (len > LOWCOMMS_MAX_TX_BUFFER_LEN) {
-		BUILD_BUG_ON(PAGE_SIZE < LOWCOMMS_MAX_TX_BUFFER_LEN);
-		log_print("failed to allocate a buffer of size %d", len);
-		return NULL;
+	spin_lock(&con->writequeue_lock);
+	if (!list_empty(&con->writequeue)) {
+		e = list_last_entry(&con->writequeue, struct writequeue_entry, list);
+		if (DLM_WQ_REMAIN_BYTES(e) >= len) {
+			*ppc = page_address(e->page) + e->end;
+			e->end += len;
+			e->users++;
+			spin_unlock(&con->writequeue_lock);
+
+			return e;
+		}
 	}
+	spin_unlock(&con->writequeue_lock);
 
-	con = nodeid2con(nodeid, allocation);
-	if (!con)
+	e = new_writequeue_entry(con, allocation);
+	if (!e)
 		return NULL;
 
+	*ppc = page_address(e->page);
+	e->end += len;
+
 	spin_lock(&con->writequeue_lock);
-	e = list_entry(con->writequeue.prev, struct writequeue_entry, list);
-	if ((&e->list == &con->writequeue) ||
-	    (PAGE_SIZE - e->end < len)) {
-		e = NULL;
-	} else {
-		offset = e->end;
-		e->end += len;
-		e->users++;
-	}
+	list_add_tail(&e->list, &con->writequeue);
 	spin_unlock(&con->writequeue_lock);
 
-	if (e) {
-	got_one:
-		*ppc = page_address(e->page) + offset;
-		return e;
-	}
+	return e;
+};
 
-	e = new_writequeue_entry(con, allocation);
-	if (e) {
-		spin_lock(&con->writequeue_lock);
-		offset = e->end;
-		e->end += len;
-		e->users++;
-		list_add_tail(&e->list, &con->writequeue);
-		spin_unlock(&con->writequeue_lock);
-		goto got_one;
+void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc)
+{
+	struct connection *con;
+
+	if (len > DEFAULT_BUFFER_SIZE ||
+	    len < sizeof(struct dlm_header)) {
+		BUILD_BUG_ON(PAGE_SIZE < DEFAULT_BUFFER_SIZE);
+		log_print("failed to allocate a buffer of size %d", len);
+		WARN_ON(1);
+		return NULL;
 	}
-	return NULL;
+
+	con = nodeid2con(nodeid, allocation);
+	if (!con)
+		return NULL;
+
+	return new_wq_entry(con, len, allocation, ppc);
 }
 
 void dlm_lowcomms_commit_buffer(void *mh)
@@ -1406,7 +1430,8 @@ void dlm_lowcomms_commit_buffer(void *mh)
 	users = --e->users;
 	if (users)
 		goto out;
-	e->len = e->end - e->offset;
+
+	e->len = DLM_WQ_LENGTH_BYTES(e);
 	spin_unlock(&con->writequeue_lock);
 
 	queue_work(send_workqueue, &con->swork);
@@ -1432,11 +1457,10 @@ static void send_to_sock(struct connection *con)
 
 	spin_lock(&con->writequeue_lock);
 	for (;;) {
-		e = list_entry(con->writequeue.next, struct writequeue_entry,
-			       list);
-		if ((struct list_head *) e == &con->writequeue)
+		if (list_empty(&con->writequeue))
 			break;
 
+		e = list_first_entry(&con->writequeue, struct writequeue_entry, list);
 		len = e->len;
 		offset = e->offset;
 		BUG_ON(len == 0 && e->users == 0);
@@ -1589,6 +1613,29 @@ static int work_start(void)
 	return 0;
 }
 
+static void shutdown_conn(struct connection *con)
+{
+	if (con->shutdown_action)
+		con->shutdown_action(con);
+}
+
+void dlm_lowcomms_shutdown(void)
+{
+	/* Set all the flags to prevent any
+	 * socket activity.
+	 */
+	dlm_allow_conn = 0;
+
+	if (recv_workqueue)
+		flush_workqueue(recv_workqueue);
+	if (send_workqueue)
+		flush_workqueue(send_workqueue);
+
+	dlm_close_sock(&listen_con.sock);
+
+	foreach_conn(shutdown_conn);
+}
+
 static void _stop_conn(struct connection *con, bool and_other)
 {
 	mutex_lock(&con->sock_mutex);
@@ -1610,12 +1657,6 @@ static void stop_conn(struct connection *con)
 	_stop_conn(con, true);
 }
 
-static void shutdown_conn(struct connection *con)
-{
-	if (con->shutdown_action)
-		con->shutdown_action(con);
-}
-
 static void connection_release(struct rcu_head *rcu)
 {
 	struct connection *con = container_of(rcu, struct connection, rcu);
@@ -1672,19 +1713,6 @@ static void work_flush(void)
 
 void dlm_lowcomms_stop(void)
 {
-	/* Set all the flags to prevent any
-	   socket activity.
-	*/
-	dlm_allow_conn = 0;
-
-	if (recv_workqueue)
-		flush_workqueue(recv_workqueue);
-	if (send_workqueue)
-		flush_workqueue(send_workqueue);
-
-	dlm_close_sock(&listen_con.sock);
-
-	foreach_conn(shutdown_conn);
 	work_flush();
 	foreach_conn(free_conn);
 	work_stop();
diff --git a/fs/dlm/lowcomms.h b/fs/dlm/lowcomms.h
index 0918f9376489..48bbc4e18761 100644
--- a/fs/dlm/lowcomms.h
+++ b/fs/dlm/lowcomms.h
@@ -14,13 +14,18 @@
 
 #define LOWCOMMS_MAX_TX_BUFFER_LEN	4096
 
+/* switch to check if dlm is running */
+extern int dlm_allow_conn;
+
 int dlm_lowcomms_start(void);
+void dlm_lowcomms_shutdown(void);
 void dlm_lowcomms_stop(void);
 void dlm_lowcomms_exit(void);
 int dlm_lowcomms_close(int nodeid);
 void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc);
 void dlm_lowcomms_commit_buffer(void *mh);
 int dlm_lowcomms_connect_node(int nodeid);
+int dlm_lowcomms_nodes_set_mark(int nodeid, unsigned int mark);
 int dlm_lowcomms_addr(int nodeid, struct sockaddr_storage *addr, int len);
 
 #endif				/* __LOWCOMMS_DOT_H__ */
diff --git a/fs/dlm/midcomms.c b/fs/dlm/midcomms.c
index fde3a6afe4be..1c6654a21ec4 100644
--- a/fs/dlm/midcomms.c
+++ b/fs/dlm/midcomms.c
@@ -22,8 +22,6 @@
  * into packets and sends them to the comms layer.
  */
 
-#include <asm/unaligned.h>
-
 #include "dlm_internal.h"
 #include "lowcomms.h"
 #include "config.h"
@@ -45,13 +43,22 @@ int dlm_process_incoming_buffer(int nodeid, unsigned char *buf, int len)
 	while (len >= sizeof(struct dlm_header)) {
 		hd = (struct dlm_header *)ptr;
 
-		/* no message should be more than this otherwise we
-		 * cannot deliver this message to upper layers
+		/* no message should be more than DEFAULT_BUFFER_SIZE or
+		 * less than dlm_header size.
+		 *
+		 * Some messages does not have a 8 byte length boundary yet
+		 * which can occur in a unaligned memory access of some dlm
+		 * messages. However this problem need to be fixed at the
+		 * sending side, for now it seems nobody run into architecture
+		 * related issues yet but it slows down some processing.
+		 * Fixing this issue should be scheduled in future by doing
+		 * the next major version bump.
 		 */
-		msglen = get_unaligned_le16(&hd->h_length);
-		if (msglen > DEFAULT_BUFFER_SIZE) {
-			log_print("received invalid length header: %u, will abort message parsing",
-				  msglen);
+		msglen = le16_to_cpu(hd->h_length);
+		if (msglen > DEFAULT_BUFFER_SIZE ||
+		    msglen < sizeof(struct dlm_header)) {
+			log_print("received invalid length header: %u from node %d, will abort message parsing",
+				  msglen, nodeid);
 			return -EBADMSG;
 		}
 
@@ -84,15 +91,7 @@ int dlm_process_incoming_buffer(int nodeid, unsigned char *buf, int len)
 			goto skip;
 		}
 
-		/* for aligned memory access, we just copy current message
-		 * to begin of the buffer which contains already parsed buffer
-		 * data and should provide align access for upper layers
-		 * because the start address of the buffer has a aligned
-		 * address. This memmove can be removed when the upperlayer
-		 * is capable of unaligned memory access.
-		 */
-		memmove(buf, ptr, msglen);
-		dlm_receive_buffer((union dlm_packet *)buf, nodeid);
+		dlm_receive_buffer((union dlm_packet *)ptr, nodeid);
 
 skip:
 		ret += msglen;
diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c
index 73ddee5159d7..f5b1bd65728d 100644
--- a/fs/dlm/rcom.c
+++ b/fs/dlm/rcom.c
@@ -41,7 +41,6 @@ static int create_rcom(struct dlm_ls *ls, int to_nodeid, int type, int len,
 			  to_nodeid, type, len);
 		return -ENOBUFS;
 	}
-	memset(mb, 0, mb_len);
 
 	rc = (struct dlm_rcom *) mb;
 
@@ -462,7 +461,6 @@ int dlm_send_ls_not_ready(int nodeid, struct dlm_rcom *rc_in)
 	mh = dlm_lowcomms_get_buffer(nodeid, mb_len, GFP_NOFS, &mb);
 	if (!mh)
 		return -ENOBUFS;
-	memset(mb, 0, mb_len);
 
 	rc = (struct dlm_rcom *) mb;