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

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)
 {
-	return !x;
+	switch (x) {
+	case 0:
+		/* TCP */
+		break;
+	case 1:
+		/* SCTP */
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (dlm_allow_conn)
+		return -EBUSY;
+
+	return 0;
 }
 
-static bool dlm_check_buffer_size(unsigned int x)
+static int dlm_check_zero_and_dlm_running(unsigned int x)
 {
-	return (x < DEFAULT_BUFFER_SIZE);
+	if (!x)
+		return -EINVAL;
+
+	if (dlm_allow_conn)
+		return -EBUSY;
+
+	return 0;
 }
 
-CLUSTER_ATTR(tcp_port, dlm_check_zero);
+static int dlm_check_zero(unsigned int x)
+{
+	if (!x)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int dlm_check_buffer_size(unsigned int x)
+{
+	if (x < DEFAULT_BUFFER_SIZE)
+		return -EINVAL;
+
+	return 0;
+}
+
+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;

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);
 

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;

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);

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;

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__ */

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;

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;