Merge tag 'md/4.3' of git://neil.brown.name/md

Pull md updates from Neil Brown:

 - an assortment of little fixes, several for minor races only likely to
   be hit during testing

 - further cluster-md-raid1 development, not ready for real use yet.

 - new RAID6 syndrome code for ARM NEON

 - fix a race where a write can return before failure of one device is
   properly recorded in metadata, so an immediate crash might result in
   that write being lost.

* tag 'md/4.3' of git://neil.brown.name/md: (33 commits)
  md/raid5: ensure device failure recorded before write request returns.
  md/raid5: use bio_list for the list of bios to return.
  md/raid10: ensure device failure recorded before write request returns.
  md/raid1: ensure device failure recorded before write request returns.
  md-cluster: remove inappropriate try_module_get from join()
  md: extend spinlock protection in register_md_cluster_operations
  md-cluster: Read the disk bitmap sb and check if it needs recovery
  md-cluster: only call complete(&cinfo->completion) when node join cluster
  md-cluster: add missed lockres_free
  md-cluster: remove the unused sb_lock
  md-cluster: init suspend_list and suspend_lock early in join
  md-cluster: add the error check if failed to get dlm lock
  md-cluster: init completion within lockres_init
  md-cluster: fix deadlock issue on message lock
  md-cluster: transfer the resync ownership to another node
  md-cluster: split recover_slot for future code reuse
  md-cluster: use %pU to print UUIDs
  md: setup safemode_timer before it's being used
  md/raid5: handle possible race as reshape completes.
  md: sync sync_completed has correct value as recovery finishes.
  ...

9 files changed, 373 insertions, 190 deletions

diff --git a/drivers/md/md-cluster.c b/drivers/md/md-cluster.c
index 0072190515e0..11e3bc9d2a4b 100644
--- a/drivers/md/md-cluster.c
+++ b/drivers/md/md-cluster.c
@@ -45,6 +45,7 @@ struct resync_info {
 /* md_cluster_info flags */
 #define		MD_CLUSTER_WAITING_FOR_NEWDISK		1
 #define		MD_CLUSTER_SUSPEND_READ_BALANCING	2
+#define		MD_CLUSTER_BEGIN_JOIN_CLUSTER		3
 
 
 struct md_cluster_info {
@@ -52,7 +53,6 @@ struct md_cluster_info {
 	dlm_lockspace_t *lockspace;
 	int slot_number;
 	struct completion completion;
-	struct dlm_lock_resource *sb_lock;
 	struct mutex sb_mutex;
 	struct dlm_lock_resource *bitmap_lockres;
 	struct list_head suspend_list;
@@ -75,6 +75,7 @@ enum msg_type {
 	NEWDISK,
 	REMOVE,
 	RE_ADD,
+	BITMAP_NEEDS_SYNC,
 };
 
 struct cluster_msg {
@@ -99,7 +100,6 @@ static int dlm_lock_sync(struct dlm_lock_resource *res, int mode)
 {
 	int ret = 0;
 
-	init_completion(&res->completion);
 	ret = dlm_lock(res->ls, mode, &res->lksb,
 			res->flags, res->name, strlen(res->name),
 			0, sync_ast, res, res->bast);
@@ -124,6 +124,7 @@ static struct dlm_lock_resource *lockres_init(struct mddev *mddev,
 	res = kzalloc(sizeof(struct dlm_lock_resource), GFP_KERNEL);
 	if (!res)
 		return NULL;
+	init_completion(&res->completion);
 	res->ls = cinfo->lockspace;
 	res->mddev = mddev;
 	namelen = strlen(name);
@@ -165,11 +166,24 @@ out_err:
 
 static void lockres_free(struct dlm_lock_resource *res)
 {
+	int ret;
+
 	if (!res)
 		return;
 
-	init_completion(&res->completion);
-	dlm_unlock(res->ls, res->lksb.sb_lkid, 0, &res->lksb, res);
+	/* cancel a lock request or a conversion request that is blocked */
+	res->flags |= DLM_LKF_CANCEL;
+retry:
+	ret = dlm_unlock(res->ls, res->lksb.sb_lkid, 0, &res->lksb, res);
+	if (unlikely(ret != 0)) {
+		pr_info("%s: failed to unlock %s return %d\n", __func__, res->name, ret);
+
+		/* if a lock conversion is cancelled, then the lock is put
+		 * back to grant queue, need to ensure it is unlocked */
+		if (ret == -DLM_ECANCEL)
+			goto retry;
+	}
+	res->flags &= ~DLM_LKF_CANCEL;
 	wait_for_completion(&res->completion);
 
 	kfree(res->name);
@@ -177,18 +191,6 @@ static void lockres_free(struct dlm_lock_resource *res)
 	kfree(res);
 }
 
-static char *pretty_uuid(char *dest, char *src)
-{
-	int i, len = 0;
-
-	for (i = 0; i < 16; i++) {
-		if (i == 4 || i == 6 || i == 8 || i == 10)
-			len += sprintf(dest + len, "-");
-		len += sprintf(dest + len, "%02x", (__u8)src[i]);
-	}
-	return dest;
-}
-
 static void add_resync_info(struct mddev *mddev, struct dlm_lock_resource *lockres,
 		sector_t lo, sector_t hi)
 {
@@ -281,16 +283,11 @@ static void recover_prep(void *arg)
 	set_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state);
 }
 
-static void recover_slot(void *arg, struct dlm_slot *slot)
+static void __recover_slot(struct mddev *mddev, int slot)
 {
-	struct mddev *mddev = arg;
 	struct md_cluster_info *cinfo = mddev->cluster_info;
 
-	pr_info("md-cluster: %s Node %d/%d down. My slot: %d. Initiating recovery.\n",
-			mddev->bitmap_info.cluster_name,
-			slot->nodeid, slot->slot,
-			cinfo->slot_number);
-	set_bit(slot->slot - 1, &cinfo->recovery_map);
+	set_bit(slot, &cinfo->recovery_map);
 	if (!cinfo->recovery_thread) {
 		cinfo->recovery_thread = md_register_thread(recover_bitmaps,
 				mddev, "recover");
@@ -302,6 +299,20 @@ static void recover_slot(void *arg, struct dlm_slot *slot)
 	md_wakeup_thread(cinfo->recovery_thread);
 }
 
+static void recover_slot(void *arg, struct dlm_slot *slot)
+{
+	struct mddev *mddev = arg;
+	struct md_cluster_info *cinfo = mddev->cluster_info;
+
+	pr_info("md-cluster: %s Node %d/%d down. My slot: %d. Initiating recovery.\n",
+			mddev->bitmap_info.cluster_name,
+			slot->nodeid, slot->slot,
+			cinfo->slot_number);
+	/* deduct one since dlm slot starts from one while the num of
+	 * cluster-md begins with 0 */
+	__recover_slot(mddev, slot->slot - 1);
+}
+
 static void recover_done(void *arg, struct dlm_slot *slots,
 		int num_slots, int our_slot,
 		uint32_t generation)
@@ -310,10 +321,17 @@ static void recover_done(void *arg, struct dlm_slot *slots,
 	struct md_cluster_info *cinfo = mddev->cluster_info;
 
 	cinfo->slot_number = our_slot;
-	complete(&cinfo->completion);
+	/* completion is only need to be complete when node join cluster,
+	 * it doesn't need to run during another node's failure */
+	if (test_bit(MD_CLUSTER_BEGIN_JOIN_CLUSTER, &cinfo->state)) {
+		complete(&cinfo->completion);
+		clear_bit(MD_CLUSTER_BEGIN_JOIN_CLUSTER, &cinfo->state);
+	}
 	clear_bit(MD_CLUSTER_SUSPEND_READ_BALANCING, &cinfo->state);
 }
 
+/* the ops is called when node join the cluster, and do lock recovery
+ * if node failure occurs */
 static const struct dlm_lockspace_ops md_ls_ops = {
 	.recover_prep = recover_prep,
 	.recover_slot = recover_slot,
@@ -388,7 +406,7 @@ static void process_add_new_disk(struct mddev *mddev, struct cluster_msg *cmsg)
 	int len;
 
 	len = snprintf(disk_uuid, 64, "DEVICE_UUID=");
-	pretty_uuid(disk_uuid + len, cmsg->uuid);
+	sprintf(disk_uuid + len, "%pU", cmsg->uuid);
 	snprintf(raid_slot, 16, "RAID_DISK=%d", cmsg->raid_slot);
 	pr_info("%s:%d Sending kobject change with %s and %s\n", __func__, __LINE__, disk_uuid, raid_slot);
 	init_completion(&cinfo->newdisk_completion);
@@ -457,6 +475,11 @@ static void process_recvd_msg(struct mddev *mddev, struct cluster_msg *msg)
 			__func__, __LINE__, msg->slot);
 		process_readd_disk(mddev, msg);
 		break;
+	case BITMAP_NEEDS_SYNC:
+		pr_info("%s: %d Received BITMAP_NEEDS_SYNC from %d\n",
+			__func__, __LINE__, msg->slot);
+		__recover_slot(mddev, msg->slot);
+		break;
 	default:
 		pr_warn("%s:%d Received unknown message from %d\n",
 			__func__, __LINE__, msg->slot);
@@ -472,6 +495,7 @@ static void recv_daemon(struct md_thread *thread)
 	struct dlm_lock_resource *ack_lockres = cinfo->ack_lockres;
 	struct dlm_lock_resource *message_lockres = cinfo->message_lockres;
 	struct cluster_msg msg;
+	int ret;
 
 	/*get CR on Message*/
 	if (dlm_lock_sync(message_lockres, DLM_LOCK_CR)) {
@@ -484,13 +508,21 @@ static void recv_daemon(struct md_thread *thread)
 	process_recvd_msg(thread->mddev, &msg);
 
 	/*release CR on ack_lockres*/
-	dlm_unlock_sync(ack_lockres);
-	/*up-convert to EX on message_lockres*/
-	dlm_lock_sync(message_lockres, DLM_LOCK_EX);
+	ret = dlm_unlock_sync(ack_lockres);
+	if (unlikely(ret != 0))
+		pr_info("unlock ack failed return %d\n", ret);
+	/*up-convert to PR on message_lockres*/
+	ret = dlm_lock_sync(message_lockres, DLM_LOCK_PR);
+	if (unlikely(ret != 0))
+		pr_info("lock PR on msg failed return %d\n", ret);
 	/*get CR on ack_lockres again*/
-	dlm_lock_sync(ack_lockres, DLM_LOCK_CR);
+	ret = dlm_lock_sync(ack_lockres, DLM_LOCK_CR);
+	if (unlikely(ret != 0))
+		pr_info("lock CR on ack failed return %d\n", ret);
 	/*release CR on message_lockres*/
-	dlm_unlock_sync(message_lockres);
+	ret = dlm_unlock_sync(message_lockres);
+	if (unlikely(ret != 0))
+		pr_info("unlock msg failed return %d\n", ret);
 }
 
 /* lock_comm()
@@ -519,7 +551,7 @@ static void unlock_comm(struct md_cluster_info *cinfo)
  * The function:
  * 1. Grabs the message lockresource in EX mode
  * 2. Copies the message to the message LVB
- * 3. Downconverts message lockresource to CR
+ * 3. Downconverts message lockresource to CW
  * 4. Upconverts ack lock resource from CR to EX. This forces the BAST on other nodes
  *    and the other nodes read the message. The thread will wait here until all other
  *    nodes have released ack lock resource.
@@ -540,12 +572,12 @@ static int __sendmsg(struct md_cluster_info *cinfo, struct cluster_msg *cmsg)
 
 	memcpy(cinfo->message_lockres->lksb.sb_lvbptr, (void *)cmsg,
 			sizeof(struct cluster_msg));
-	/*down-convert EX to CR on Message*/
-	error = dlm_lock_sync(cinfo->message_lockres, DLM_LOCK_CR);
+	/*down-convert EX to CW on Message*/
+	error = dlm_lock_sync(cinfo->message_lockres, DLM_LOCK_CW);
 	if (error) {
-		pr_err("md-cluster: failed to convert EX to CR on MESSAGE(%d)\n",
+		pr_err("md-cluster: failed to convert EX to CW on MESSAGE(%d)\n",
 				error);
-		goto failed_message;
+		goto failed_ack;
 	}
 
 	/*up-convert CR to EX on Ack*/
@@ -565,7 +597,13 @@ static int __sendmsg(struct md_cluster_info *cinfo, struct cluster_msg *cmsg)
 	}
 
 failed_ack:
-	dlm_unlock_sync(cinfo->message_lockres);
+	error = dlm_unlock_sync(cinfo->message_lockres);
+	if (unlikely(error != 0)) {
+		pr_err("md-cluster: failed convert to NL on MESSAGE(%d)\n",
+			error);
+		/* in case the message can't be released due to some reason */
+		goto failed_ack;
+	}
 failed_message:
 	return error;
 }
@@ -587,6 +625,7 @@ static int gather_all_resync_info(struct mddev *mddev, int total_slots)
 	struct dlm_lock_resource *bm_lockres;
 	struct suspend_info *s;
 	char str[64];
+	sector_t lo, hi;
 
 
 	for (i = 0; i < total_slots; i++) {
@@ -617,9 +656,24 @@ static int gather_all_resync_info(struct mddev *mddev, int total_slots)
 			lockres_free(bm_lockres);
 			continue;
 		}
-		if (ret)
+		if (ret) {
+			lockres_free(bm_lockres);
 			goto out;
-		/* TODO: Read the disk bitmap sb and check if it needs recovery */
+		}
+
+		/* Read the disk bitmap sb and check if it needs recovery */
+		ret = bitmap_copy_from_slot(mddev, i, &lo, &hi, false);
+		if (ret) {
+			pr_warn("md-cluster: Could not gather bitmaps from slot %d", i);
+			lockres_free(bm_lockres);
+			continue;
+		}
+		if ((hi > 0) && (lo < mddev->recovery_cp)) {
+			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+			mddev->recovery_cp = lo;
+			md_check_recovery(mddev);
+		}
+
 		dlm_unlock_sync(bm_lockres);
 		lockres_free(bm_lockres);
 	}
@@ -633,20 +687,20 @@ static int join(struct mddev *mddev, int nodes)
 	int ret, ops_rv;
 	char str[64];
 
-	if (!try_module_get(THIS_MODULE))
-		return -ENOENT;
-
 	cinfo = kzalloc(sizeof(struct md_cluster_info), GFP_KERNEL);
 	if (!cinfo)
 		return -ENOMEM;
 
+	INIT_LIST_HEAD(&cinfo->suspend_list);
+	spin_lock_init(&cinfo->suspend_lock);
 	init_completion(&cinfo->completion);
+	set_bit(MD_CLUSTER_BEGIN_JOIN_CLUSTER, &cinfo->state);
 
 	mutex_init(&cinfo->sb_mutex);
 	mddev->cluster_info = cinfo;
 
 	memset(str, 0, 64);
-	pretty_uuid(str, mddev->uuid);
+	sprintf(str, "%pU", mddev->uuid);
 	ret = dlm_new_lockspace(str, mddev->bitmap_info.cluster_name,
 				DLM_LSFL_FS, LVB_SIZE,
 				&md_ls_ops, mddev, &ops_rv, &cinfo->lockspace);
@@ -659,12 +713,6 @@ static int join(struct mddev *mddev, int nodes)
 		ret = -ERANGE;
 		goto err;
 	}
-	cinfo->sb_lock = lockres_init(mddev, "cmd-super",
-					NULL, 0);
-	if (!cinfo->sb_lock) {
-		ret = -ENOMEM;
-		goto err;
-	}
 	/* Initiate the communication resources */
 	ret = -ENOMEM;
 	cinfo->recv_thread = md_register_thread(recv_daemon, mddev, "cluster_recv");
@@ -705,9 +753,6 @@ static int join(struct mddev *mddev, int nodes)
 		goto err;
 	}
 
-	INIT_LIST_HEAD(&cinfo->suspend_list);
-	spin_lock_init(&cinfo->suspend_lock);
-
 	ret = gather_all_resync_info(mddev, nodes);
 	if (ret)
 		goto err;
@@ -719,12 +764,10 @@ err:
 	lockres_free(cinfo->ack_lockres);
 	lockres_free(cinfo->no_new_dev_lockres);
 	lockres_free(cinfo->bitmap_lockres);
-	lockres_free(cinfo->sb_lock);
 	if (cinfo->lockspace)
 		dlm_release_lockspace(cinfo->lockspace, 2);
 	mddev->cluster_info = NULL;
 	kfree(cinfo);
-	module_put(THIS_MODULE);
 	return ret;
 }
 
@@ -740,7 +783,6 @@ static int leave(struct mddev *mddev)
 	lockres_free(cinfo->token_lockres);
 	lockres_free(cinfo->ack_lockres);
 	lockres_free(cinfo->no_new_dev_lockres);
-	lockres_free(cinfo->sb_lock);
 	lockres_free(cinfo->bitmap_lockres);
 	dlm_release_lockspace(cinfo->lockspace, 2);
 	return 0;
@@ -817,8 +859,17 @@ static int resync_start(struct mddev *mddev, sector_t lo, sector_t hi)
 
 static void resync_finish(struct mddev *mddev)
 {
+	struct md_cluster_info *cinfo = mddev->cluster_info;
+	struct cluster_msg cmsg;
+	int slot = cinfo->slot_number - 1;
+
 	pr_info("%s:%d\n", __func__, __LINE__);
 	resync_send(mddev, RESYNCING, 0, 0);
+	if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
+		cmsg.type = cpu_to_le32(BITMAP_NEEDS_SYNC);
+		cmsg.slot = cpu_to_le32(slot);
+		sendmsg(cinfo, &cmsg);
+	}
 }
 
 static int area_resyncing(struct mddev *mddev, int direction,
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 40332625713b..4f5ecbe94ccb 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -483,6 +483,8 @@ static void mddev_put(struct mddev *mddev)
 		bioset_free(bs);
 }
 
+static void md_safemode_timeout(unsigned long data);
+
 void mddev_init(struct mddev *mddev)
 {
 	mutex_init(&mddev->open_mutex);
@@ -490,7 +492,8 @@ void mddev_init(struct mddev *mddev)
 	mutex_init(&mddev->bitmap_info.mutex);
 	INIT_LIST_HEAD(&mddev->disks);
 	INIT_LIST_HEAD(&mddev->all_mddevs);
-	init_timer(&mddev->safemode_timer);
+	setup_timer(&mddev->safemode_timer, md_safemode_timeout,
+		    (unsigned long) mddev);
 	atomic_set(&mddev->active, 1);
 	atomic_set(&mddev->openers, 0);
 	atomic_set(&mddev->active_io, 0);
@@ -3255,8 +3258,6 @@ int strict_strtoul_scaled(const char *cp, unsigned long *res, int scale)
 	return 0;
 }
 
-static void md_safemode_timeout(unsigned long data);
-
 static ssize_t
 safe_delay_show(struct mddev *mddev, char *page)
 {
@@ -4189,6 +4190,8 @@ action_show(struct mddev *mddev, char *page)
 				type = "repair";
 		} else if (test_bit(MD_RECOVERY_RECOVER, &recovery))
 			type = "recover";
+		else if (mddev->reshape_position != MaxSector)
+			type = "reshape";
 	}
 	return sprintf(page, "%s\n", type);
 }
@@ -5180,8 +5183,6 @@ int md_run(struct mddev *mddev)
 	atomic_set(&mddev->max_corr_read_errors,
 		   MD_DEFAULT_MAX_CORRECTED_READ_ERRORS);
 	mddev->safemode = 0;
-	mddev->safemode_timer.function = md_safemode_timeout;
-	mddev->safemode_timer.data = (unsigned long) mddev;
 	mddev->safemode_delay = (200 * HZ)/1000 +1; /* 200 msec delay */
 	mddev->in_sync = 1;
 	smp_wmb();
@@ -5194,6 +5195,11 @@ int md_run(struct mddev *mddev)
 			if (sysfs_link_rdev(mddev, rdev))
 				/* failure here is OK */;
 
+	if (mddev->degraded && !mddev->ro)
+		/* This ensures that recovering status is reported immediately
+		 * via sysfs - until a lack of spares is confirmed.
+		 */
+		set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 
 	if (mddev->flags & MD_UPDATE_SB_FLAGS)
@@ -5741,16 +5747,16 @@ static int get_bitmap_file(struct mddev *mddev, void __user * arg)
 
 	err = 0;
 	spin_lock(&mddev->lock);
-	/* bitmap disabled, zero the first byte and copy out */
-	if (!mddev->bitmap_info.file)
-		file->pathname[0] = '\0';
-	else if ((ptr = file_path(mddev->bitmap_info.file,
-			       file->pathname, sizeof(file->pathname))),
-		 IS_ERR(ptr))
-		err = PTR_ERR(ptr);
-	else
-		memmove(file->pathname, ptr,
-			sizeof(file->pathname)-(ptr-file->pathname));
+	/* bitmap enabled */
+	if (mddev->bitmap_info.file) {
+		ptr = file_path(mddev->bitmap_info.file, file->pathname,
+				sizeof(file->pathname));
+		if (IS_ERR(ptr))
+			err = PTR_ERR(ptr);
+		else
+			memmove(file->pathname, ptr,
+				sizeof(file->pathname)-(ptr-file->pathname));
+	}
 	spin_unlock(&mddev->lock);
 
 	if (err == 0 &&
@@ -7069,7 +7075,7 @@ static void status_unused(struct seq_file *seq)
 	seq_printf(seq, "\n");
 }
 
-static void status_resync(struct seq_file *seq, struct mddev *mddev)
+static int status_resync(struct seq_file *seq, struct mddev *mddev)
 {
 	sector_t max_sectors, resync, res;
 	unsigned long dt, db;
@@ -7077,18 +7083,32 @@ static void status_resync(struct seq_file *seq, struct mddev *mddev)
 	int scale;
 	unsigned int per_milli;
 
-	if (mddev->curr_resync <= 3)
-		resync = 0;
-	else
-		resync = mddev->curr_resync
-			- atomic_read(&mddev->recovery_active);
-
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ||
 	    test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
 		max_sectors = mddev->resync_max_sectors;
 	else
 		max_sectors = mddev->dev_sectors;
 
+	resync = mddev->curr_resync;
+	if (resync <= 3) {
+		if (test_bit(MD_RECOVERY_DONE, &mddev->recovery))
+			/* Still cleaning up */
+			resync = max_sectors;
+	} else
+		resync -= atomic_read(&mddev->recovery_active);
+
+	if (resync == 0) {
+		if (mddev->recovery_cp < MaxSector) {
+			seq_printf(seq, "\tresync=PENDING");
+			return 1;
+		}
+		return 0;
+	}
+	if (resync < 3) {
+		seq_printf(seq, "\tresync=DELAYED");
+		return 1;
+	}
+
 	WARN_ON(max_sectors == 0);
 	/* Pick 'scale' such that (resync>>scale)*1000 will fit
 	 * in a sector_t, and (max_sectors>>scale) will fit in a
@@ -7153,6 +7173,7 @@ static void status_resync(struct seq_file *seq, struct mddev *mddev)
 		   ((unsigned long)rt % 60)/6);
 
 	seq_printf(seq, " speed=%ldK/sec", db/2/dt);
+	return 1;
 }
 
 static void *md_seq_start(struct seq_file *seq, loff_t *pos)
@@ -7298,13 +7319,8 @@ static int md_seq_show(struct seq_file *seq, void *v)
 			mddev->pers->status(seq, mddev);
 			seq_printf(seq, "\n      ");
 			if (mddev->pers->sync_request) {
-				if (mddev->curr_resync > 2) {
-					status_resync(seq, mddev);
+				if (status_resync(seq, mddev))
 					seq_printf(seq, "\n      ");
-				} else if (mddev->curr_resync >= 1)
-					seq_printf(seq, "\tresync=DELAYED\n      ");
-				else if (mddev->recovery_cp < MaxSector)
-					seq_printf(seq, "\tresync=PENDING\n      ");
 			}
 		} else
 			seq_printf(seq, "\n       ");
@@ -7387,15 +7403,19 @@ int unregister_md_personality(struct md_personality *p)
 }
 EXPORT_SYMBOL(unregister_md_personality);
 
-int register_md_cluster_operations(struct md_cluster_operations *ops, struct module *module)
+int register_md_cluster_operations(struct md_cluster_operations *ops,
+				   struct module *module)
 {
-	if (md_cluster_ops != NULL)
-		return -EALREADY;
+	int ret = 0;
 	spin_lock(&pers_lock);
-	md_cluster_ops = ops;
-	md_cluster_mod = module;
+	if (md_cluster_ops != NULL)
+		ret = -EALREADY;
+	else {
+		md_cluster_ops = ops;
+		md_cluster_mod = module;
+	}
 	spin_unlock(&pers_lock);
-	return 0;
+	return ret;
 }
 EXPORT_SYMBOL(register_md_cluster_operations);
 
@@ -7793,7 +7813,8 @@ void md_do_sync(struct md_thread *thread)
 		      > (max_sectors >> 4)) ||
 		     time_after_eq(jiffies, update_time + UPDATE_FREQUENCY) ||
 		     (j - mddev->curr_resync_completed)*2
-		     >= mddev->resync_max - mddev->curr_resync_completed
+		     >= mddev->resync_max - mddev->curr_resync_completed ||
+		     mddev->curr_resync_completed > mddev->resync_max
 			    )) {
 			/* time to update curr_resync_completed */
 			wait_event(mddev->recovery_wait,
@@ -7838,6 +7859,9 @@ void md_do_sync(struct md_thread *thread)
 			break;
 
 		j += sectors;
+		if (j > max_sectors)
+			/* when skipping, extra large numbers can be returned. */
+			j = max_sectors;
 		if (j > 2)
 			mddev->curr_resync = j;
 		if (mddev_is_clustered(mddev))
@@ -7906,12 +7930,15 @@ void md_do_sync(struct md_thread *thread)
 	blk_finish_plug(&plug);
 	wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active));
 
+	if (!test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
+	    !test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
+	    mddev->curr_resync > 2) {
+		mddev->curr_resync_completed = mddev->curr_resync;
+		sysfs_notify(&mddev->kobj, NULL, "sync_completed");
+	}
 	/* tell personality that we are finished */
 	mddev->pers->sync_request(mddev, max_sectors, &skipped);
 
-	if (mddev_is_clustered(mddev))
-		md_cluster_ops->resync_finish(mddev);
-
 	if (!test_bit(MD_RECOVERY_CHECK, &mddev->recovery) &&
 	    mddev->curr_resync > 2) {
 		if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
@@ -7945,6 +7972,9 @@ void md_do_sync(struct md_thread *thread)
 		}
 	}
  skip:
+	if (mddev_is_clustered(mddev))
+		md_cluster_ops->resync_finish(mddev);
+
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
 
 	spin_lock(&mddev->lock);
@@ -7955,11 +7985,11 @@ void md_do_sync(struct md_thread *thread)
 		mddev->resync_max = MaxSector;
 	} else if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
 		mddev->resync_min = mddev->curr_resync_completed;
+	set_bit(MD_RECOVERY_DONE, &mddev->recovery);
 	mddev->curr_resync = 0;
 	spin_unlock(&mddev->lock);
 
 	wake_up(&resync_wait);
-	set_bit(MD_RECOVERY_DONE, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
 	return;
 }
@@ -8128,6 +8158,7 @@ void md_check_recovery(struct mddev *mddev)
 			 */
 			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 			md_reap_sync_thread(mddev);
+			clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 			clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 			goto unlock;
 		}
@@ -8574,6 +8605,7 @@ int rdev_set_badblocks(struct md_rdev *rdev, sector_t s, int sectors,
 		/* Make sure they get written out promptly */
 		sysfs_notify_dirent_safe(rdev->sysfs_state);
 		set_bit(MD_CHANGE_CLEAN, &rdev->mddev->flags);
+		set_bit(MD_CHANGE_PENDING, &rdev->mddev->flags);
 		md_wakeup_thread(rdev->mddev->thread);
 	}
 	return rv;
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 59cda501a224..63e619b2f44e 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -83,7 +83,7 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 	char b[BDEVNAME_SIZE];
 	char b2[BDEVNAME_SIZE];
 	struct r0conf *conf = kzalloc(sizeof(*conf), GFP_KERNEL);
-	bool discard_supported = false;
+	unsigned short blksize = 512;
 
 	if (!conf)
 		return -ENOMEM;
@@ -98,6 +98,9 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 		sector_div(sectors, mddev->chunk_sectors);
 		rdev1->sectors = sectors * mddev->chunk_sectors;
 
+		blksize = max(blksize, queue_logical_block_size(
+				      rdev1->bdev->bd_disk->queue));
+
 		rdev_for_each(rdev2, mddev) {
 			pr_debug("md/raid0:%s:   comparing %s(%llu)"
 				 " with %s(%llu)\n",
@@ -134,6 +137,18 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 	}
 	pr_debug("md/raid0:%s: FINAL %d zones\n",
 		 mdname(mddev), conf->nr_strip_zones);
+	/*
+	 * now since we have the hard sector sizes, we can make sure
+	 * chunk size is a multiple of that sector size
+	 */
+	if ((mddev->chunk_sectors << 9) % blksize) {
+		printk(KERN_ERR "md/raid0:%s: chunk_size of %d not multiple of block size %d\n",
+		       mdname(mddev),
+		       mddev->chunk_sectors << 9, blksize);
+		err = -EINVAL;
+		goto abort;
+	}
+
 	err = -ENOMEM;
 	conf->strip_zone = kzalloc(sizeof(struct strip_zone)*
 				conf->nr_strip_zones, GFP_KERNEL);
@@ -188,16 +203,9 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 		}
 		dev[j] = rdev1;
 
-		if (mddev->queue)
-			disk_stack_limits(mddev->gendisk, rdev1->bdev,
-					  rdev1->data_offset << 9);
-
 		if (!smallest || (rdev1->sectors < smallest->sectors))
 			smallest = rdev1;
 		cnt++;
-
-		if (blk_queue_discard(bdev_get_queue(rdev1->bdev)))
-			discard_supported = true;
 	}
 	if (cnt != mddev->raid_disks) {
 		printk(KERN_ERR "md/raid0:%s: too few disks (%d of %d) - "
@@ -258,28 +266,6 @@ static int create_strip_zones(struct mddev *mddev, struct r0conf **private_conf)
 			 (unsigned long long)smallest->sectors);
 	}
 
-	/*
-	 * now since we have the hard sector sizes, we can make sure
-	 * chunk size is a multiple of that sector size
-	 */
-	if ((mddev->chunk_sectors << 9) % queue_logical_block_size(mddev->queue)) {
-		printk(KERN_ERR "md/raid0:%s: chunk_size of %d not valid\n",
-		       mdname(mddev),
-		       mddev->chunk_sectors << 9);
-		goto abort;
-	}
-
-	if (mddev->queue) {
-		blk_queue_io_min(mddev->queue, mddev->chunk_sectors << 9);
-		blk_queue_io_opt(mddev->queue,
-				 (mddev->chunk_sectors << 9) * mddev->raid_disks);
-
-		if (!discard_supported)
-			queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
-		else
-			queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
-	}
-
 	pr_debug("md/raid0:%s: done.\n", mdname(mddev));
 	*private_conf = conf;
 
@@ -378,12 +364,6 @@ static int raid0_run(struct mddev *mddev)
 	if (md_check_no_bitmap(mddev))
 		return -EINVAL;
 
-	if (mddev->queue) {
-		blk_queue_max_hw_sectors(mddev->queue, mddev->chunk_sectors);
-		blk_queue_max_write_same_sectors(mddev->queue, mddev->chunk_sectors);
-		blk_queue_max_discard_sectors(mddev->queue, mddev->chunk_sectors);
-	}
-
 	/* if private is not null, we are here after takeover */
 	if (mddev->private == NULL) {
 		ret = create_strip_zones(mddev, &conf);
@@ -392,6 +372,29 @@ static int raid0_run(struct mddev *mddev)
 		mddev->private = conf;
 	}
 	conf = mddev->private;
+	if (mddev->queue) {
+		struct md_rdev *rdev;
+		bool discard_supported = false;
+
+		rdev_for_each(rdev, mddev) {
+			disk_stack_limits(mddev->gendisk, rdev->bdev,
+					  rdev->data_offset << 9);
+			if (blk_queue_discard(bdev_get_queue(rdev->bdev)))
+				discard_supported = true;
+		}
+		blk_queue_max_hw_sectors(mddev->queue, mddev->chunk_sectors);
+		blk_queue_max_write_same_sectors(mddev->queue, mddev->chunk_sectors);
+		blk_queue_max_discard_sectors(mddev->queue, mddev->chunk_sectors);
+
+		blk_queue_io_min(mddev->queue, mddev->chunk_sectors << 9);
+		blk_queue_io_opt(mddev->queue,
+				 (mddev->chunk_sectors << 9) * mddev->raid_disks);
+
+		if (!discard_supported)
+			queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+		else
+			queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);
+	}
 
 	/* calculate array device size */
 	md_set_array_sectors(mddev, raid0_size(mddev, 0, 0));
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index f39d69f884de..4517f06c41ba 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1474,6 +1474,7 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
 	 */
 	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
+	set_bit(MD_CHANGE_PENDING, &mddev->flags);
 	printk(KERN_ALERT
 	       "md/raid1:%s: Disk failure on %s, disabling device.\n"
 	       "md/raid1:%s: Operation continuing on %d devices.\n",
@@ -2235,6 +2236,7 @@ static void handle_sync_write_finished(struct r1conf *conf, struct r1bio *r1_bio
 static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 {
 	int m;
+	bool fail = false;
 	for (m = 0; m < conf->raid_disks * 2 ; m++)
 		if (r1_bio->bios[m] == IO_MADE_GOOD) {
 			struct md_rdev *rdev = conf->mirrors[m].rdev;
@@ -2247,6 +2249,7 @@ static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 			 * narrow down and record precise write
 			 * errors.
 			 */
+			fail = true;
 			if (!narrow_write_error(r1_bio, m)) {
 				md_error(conf->mddev,
 					 conf->mirrors[m].rdev);
@@ -2258,7 +2261,13 @@ static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 		}
 	if (test_bit(R1BIO_WriteError, &r1_bio->state))
 		close_write(r1_bio);
-	raid_end_bio_io(r1_bio);
+	if (fail) {
+		spin_lock_irq(&conf->device_lock);
+		list_add(&r1_bio->retry_list, &conf->bio_end_io_list);
+		spin_unlock_irq(&conf->device_lock);
+		md_wakeup_thread(conf->mddev->thread);
+	} else
+		raid_end_bio_io(r1_bio);
 }
 
 static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio)
@@ -2364,6 +2373,23 @@ static void raid1d(struct md_thread *thread)
 
 	md_check_recovery(mddev);
 
+	if (!list_empty_careful(&conf->bio_end_io_list) &&
+	    !test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+		LIST_HEAD(tmp);
+		spin_lock_irqsave(&conf->device_lock, flags);
+		if (!test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+			list_add(&tmp, &conf->bio_end_io_list);
+			list_del_init(&conf->bio_end_io_list);
+		}
+		spin_unlock_irqrestore(&conf->device_lock, flags);
+		while (!list_empty(&tmp)) {
+			r1_bio = list_first_entry(&conf->bio_end_io_list,
+						  struct r1bio, retry_list);
+			list_del(&r1_bio->retry_list);
+			raid_end_bio_io(r1_bio);
+		}
+	}
+
 	blk_start_plug(&plug);
 	for (;;) {
 
@@ -2763,6 +2789,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 	conf->raid_disks = mddev->raid_disks;
 	conf->mddev = mddev;
 	INIT_LIST_HEAD(&conf->retry_list);
+	INIT_LIST_HEAD(&conf->bio_end_io_list);
 
 	spin_lock_init(&conf->resync_lock);
 	init_waitqueue_head(&conf->wait_barrier);
@@ -3057,6 +3084,7 @@ static int raid1_reshape(struct mddev *mddev)
 
 	unfreeze_array(conf);
 
+	set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
 
diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h
index 14ebb288c1ef..c52d7139c5d7 100644
--- a/drivers/md/raid1.h
+++ b/drivers/md/raid1.h
@@ -61,6 +61,11 @@ struct r1conf {
 	 * block, or anything else.
 	 */
 	struct list_head	retry_list;
+	/* A separate list of r1bio which just need raid_end_bio_io called.
+	 * This mustn't happen for writes which had any errors if the superblock
+	 * needs to be written.
+	 */
+	struct list_head	bio_end_io_list;
 
 	/* queue pending writes to be submitted on unplug */
 	struct bio_list		pending_bio_list;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index b0fce2ebf7ad..0fc33eb88855 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1589,6 +1589,7 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
 	set_bit(Blocked, &rdev->flags);
 	set_bit(Faulty, &rdev->flags);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
+	set_bit(MD_CHANGE_PENDING, &mddev->flags);
 	spin_unlock_irqrestore(&conf->device_lock, flags);
 	printk(KERN_ALERT
 	       "md/raid10:%s: Disk failure on %s, disabling device.\n"
@@ -2623,6 +2624,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 		}
 		put_buf(r10_bio);
 	} else {
+		bool fail = false;
 		for (m = 0; m < conf->copies; m++) {
 			int dev = r10_bio->devs[m].devnum;
 			struct bio *bio = r10_bio->devs[m].bio;
@@ -2634,6 +2636,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 					r10_bio->sectors, 0);
 				rdev_dec_pending(rdev, conf->mddev);
 			} else if (bio != NULL && bio->bi_error) {
+				fail = true;
 				if (!narrow_write_error(r10_bio, m)) {
 					md_error(conf->mddev, rdev);
 					set_bit(R10BIO_Degraded,
@@ -2654,7 +2657,13 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio)
 		if (test_bit(R10BIO_WriteError,
 			     &r10_bio->state))
 			close_write(r10_bio);
-		raid_end_bio_io(r10_bio);
+		if (fail) {
+			spin_lock_irq(&conf->device_lock);
+			list_add(&r10_bio->retry_list, &conf->bio_end_io_list);
+			spin_unlock_irq(&conf->device_lock);
+			md_wakeup_thread(conf->mddev->thread);
+		} else
+			raid_end_bio_io(r10_bio);
 	}
 }
 
@@ -2669,6 +2678,23 @@ static void raid10d(struct md_thread *thread)
 
 	md_check_recovery(mddev);
 
+	if (!list_empty_careful(&conf->bio_end_io_list) &&
+	    !test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+		LIST_HEAD(tmp);
+		spin_lock_irqsave(&conf->device_lock, flags);
+		if (!test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+			list_add(&tmp, &conf->bio_end_io_list);
+			list_del_init(&conf->bio_end_io_list);
+		}
+		spin_unlock_irqrestore(&conf->device_lock, flags);
+		while (!list_empty(&tmp)) {
+			r10_bio = list_first_entry(&conf->bio_end_io_list,
+						  struct r10bio, retry_list);
+			list_del(&r10_bio->retry_list);
+			raid_end_bio_io(r10_bio);
+		}
+	}
+
 	blk_start_plug(&plug);
 	for (;;) {
 
@@ -3443,6 +3469,7 @@ static struct r10conf *setup_conf(struct mddev *mddev)
 	conf->reshape_safe = conf->reshape_progress;
 	spin_lock_init(&conf->device_lock);
 	INIT_LIST_HEAD(&conf->retry_list);
+	INIT_LIST_HEAD(&conf->bio_end_io_list);
 
 	spin_lock_init(&conf->resync_lock);
 	init_waitqueue_head(&conf->wait_barrier);
@@ -4097,7 +4124,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
 	 * at a time, possibly less if that exceeds RESYNC_PAGES,
 	 * or we hit a bad block or something.
 	 * This might mean we pause for normal IO in the middle of
-	 * a chunk, but that is not a problem was mddev->reshape_position
+	 * a chunk, but that is not a problem as mddev->reshape_position
 	 * can record any location.
 	 *
 	 * If we will want to write to a location that isn't
@@ -4121,7 +4148,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
 	 *
 	 * In all this the minimum difference in data offsets
 	 * (conf->offset_diff - always positive) allows a bit of slack,
-	 * so next can be after 'safe', but not by more than offset_disk
+	 * so next can be after 'safe', but not by more than offset_diff
 	 *
 	 * We need to prepare all the bios here before we start any IO
 	 * to ensure the size we choose is acceptable to all devices.
diff --git a/drivers/md/raid10.h b/drivers/md/raid10.h
index 5ee6473ddc2c..6fc2c75759bf 100644
--- a/drivers/md/raid10.h
+++ b/drivers/md/raid10.h
@@ -53,6 +53,12 @@ struct r10conf {
 	sector_t		offset_diff;
 
 	struct list_head	retry_list;
+	/* A separate list of r1bio which just need raid_end_bio_io called.
+	 * This mustn't happen for writes which had any errors if the superblock
+	 * needs to be written.
+	 */
+	struct list_head	bio_end_io_list;
+
 	/* queue pending writes and submit them on unplug */
 	struct bio_list		pending_bio_list;
 	int			pending_count;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index b29e89cb815b..15ef2c641b2b 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -223,18 +223,14 @@ static int raid6_idx_to_slot(int idx, struct stripe_head *sh,
 	return slot;
 }
 
-static void return_io(struct bio *return_bi)
+static void return_io(struct bio_list *return_bi)
 {
-	struct bio *bi = return_bi;
-	while (bi) {
-
-		return_bi = bi->bi_next;
-		bi->bi_next = NULL;
+	struct bio *bi;
+	while ((bi = bio_list_pop(return_bi)) != NULL) {
 		bi->bi_iter.bi_size = 0;
 		trace_block_bio_complete(bdev_get_queue(bi->bi_bdev),
 					 bi, 0);
 		bio_endio(bi);
-		bi = return_bi;
 	}
 }
 
@@ -1177,7 +1173,7 @@ async_copy_data(int frombio, struct bio *bio, struct page **page,
 static void ops_complete_biofill(void *stripe_head_ref)
 {
 	struct stripe_head *sh = stripe_head_ref;
-	struct bio *return_bi = NULL;
+	struct bio_list return_bi = BIO_EMPTY_LIST;
 	int i;
 
 	pr_debug("%s: stripe %llu\n", __func__,
@@ -1201,17 +1197,15 @@ static void ops_complete_biofill(void *stripe_head_ref)
 			while (rbi && rbi->bi_iter.bi_sector <
 				dev->sector + STRIPE_SECTORS) {
 				rbi2 = r5_next_bio(rbi, dev->sector);
-				if (!raid5_dec_bi_active_stripes(rbi)) {
-					rbi->bi_next = return_bi;
-					return_bi = rbi;
-				}
+				if (!raid5_dec_bi_active_stripes(rbi))
+					bio_list_add(&return_bi, rbi);
 				rbi = rbi2;
 			}
 		}
 	}
 	clear_bit(STRIPE_BIOFILL_RUN, &sh->state);
 
-	return_io(return_bi);
+	return_io(&return_bi);
 
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
@@ -2517,6 +2511,7 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
 	set_bit(Blocked, &rdev->flags);
 	set_bit(Faulty, &rdev->flags);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
+	set_bit(MD_CHANGE_PENDING, &mddev->flags);
 	printk(KERN_ALERT
 	       "md/raid:%s: Disk failure on %s, disabling device.\n"
 	       "md/raid:%s: Operation continuing on %d devices.\n",
@@ -3069,7 +3064,7 @@ static void stripe_set_idx(sector_t stripe, struct r5conf *conf, int previous,
 static void
 handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 				struct stripe_head_state *s, int disks,
-				struct bio **return_bi)
+				struct bio_list *return_bi)
 {
 	int i;
 	BUG_ON(sh->batch_head);
@@ -3114,8 +3109,7 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 			bi->bi_error = -EIO;
 			if (!raid5_dec_bi_active_stripes(bi)) {
 				md_write_end(conf->mddev);
-				bi->bi_next = *return_bi;
-				*return_bi = bi;
+				bio_list_add(return_bi, bi);
 			}
 			bi = nextbi;
 		}
@@ -3139,8 +3133,7 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 			bi->bi_error = -EIO;
 			if (!raid5_dec_bi_active_stripes(bi)) {
 				md_write_end(conf->mddev);
-				bi->bi_next = *return_bi;
-				*return_bi = bi;
+				bio_list_add(return_bi, bi);
 			}
 			bi = bi2;
 		}
@@ -3163,10 +3156,8 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 					r5_next_bio(bi, sh->dev[i].sector);
 
 				bi->bi_error = -EIO;
-				if (!raid5_dec_bi_active_stripes(bi)) {
-					bi->bi_next = *return_bi;
-					*return_bi = bi;
-				}
+				if (!raid5_dec_bi_active_stripes(bi))
+					bio_list_add(return_bi, bi);
 				bi = nextbi;
 			}
 		}
@@ -3445,7 +3436,7 @@ static void break_stripe_batch_list(struct stripe_head *head_sh,
  * never LOCKED, so we don't need to test 'failed' directly.
  */
 static void handle_stripe_clean_event(struct r5conf *conf,
-	struct stripe_head *sh, int disks, struct bio **return_bi)
+	struct stripe_head *sh, int disks, struct bio_list *return_bi)
 {
 	int i;
 	struct r5dev *dev;
@@ -3479,8 +3470,7 @@ returnbi:
 					wbi2 = r5_next_bio(wbi, dev->sector);
 					if (!raid5_dec_bi_active_stripes(wbi)) {
 						md_write_end(conf->mddev);
-						wbi->bi_next = *return_bi;
-						*return_bi = wbi;
+						bio_list_add(return_bi, wbi);
 					}
 					wbi = wbi2;
 				}
@@ -4613,7 +4603,15 @@ finish:
 			md_wakeup_thread(conf->mddev->thread);
 	}
 
-	return_io(s.return_bi);
+	if (!bio_list_empty(&s.return_bi)) {
+		if (test_bit(MD_CHANGE_PENDING, &conf->mddev->flags)) {
+			spin_lock_irq(&conf->device_lock);
+			bio_list_merge(&conf->return_bi, &s.return_bi);
+			spin_unlock_irq(&conf->device_lock);
+			md_wakeup_thread(conf->mddev->thread);
+		} else
+			return_io(&s.return_bi);
+	}
 
 	clear_bit_unlock(STRIPE_ACTIVE, &sh->state);
 }
@@ -4672,12 +4670,12 @@ static int raid5_congested(struct mddev *mddev, int bits)
 
 static int in_chunk_boundary(struct mddev *mddev, struct bio *bio)
 {
+	struct r5conf *conf = mddev->private;
 	sector_t sector = bio->bi_iter.bi_sector + get_start_sect(bio->bi_bdev);
-	unsigned int chunk_sectors = mddev->chunk_sectors;
+	unsigned int chunk_sectors;
 	unsigned int bio_sectors = bio_sectors(bio);
 
-	if (mddev->new_chunk_sectors < mddev->chunk_sectors)
-		chunk_sectors = mddev->new_chunk_sectors;
+	chunk_sectors = min(conf->chunk_sectors, conf->prev_chunk_sectors);
 	return  chunk_sectors >=
 		((sector & (chunk_sectors - 1)) + bio_sectors);
 }
@@ -5325,6 +5323,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	sector_t stripe_addr;
 	int reshape_sectors;
 	struct list_head stripes;
+	sector_t retn;
 
 	if (sector_nr == 0) {
 		/* If restarting in the middle, skip the initial sectors */
@@ -5332,6 +5331,10 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 		    conf->reshape_progress < raid5_size(mddev, 0, 0)) {
 			sector_nr = raid5_size(mddev, 0, 0)
 				- conf->reshape_progress;
+		} else if (mddev->reshape_backwards &&
+			   conf->reshape_progress == MaxSector) {
+			/* shouldn't happen, but just in case, finish up.*/
+			sector_nr = MaxSector;
 		} else if (!mddev->reshape_backwards &&
 			   conf->reshape_progress > 0)
 			sector_nr = conf->reshape_progress;
@@ -5340,7 +5343,8 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 			mddev->curr_resync_completed = sector_nr;
 			sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 			*skipped = 1;
-			return sector_nr;
+			retn = sector_nr;
+			goto finish;
 		}
 	}
 
@@ -5348,10 +5352,8 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	 * If old and new chunk sizes differ, we need to process the
 	 * largest of these
 	 */
-	if (mddev->new_chunk_sectors > mddev->chunk_sectors)
-		reshape_sectors = mddev->new_chunk_sectors;
-	else
-		reshape_sectors = mddev->chunk_sectors;
+
+	reshape_sectors = max(conf->chunk_sectors, conf->prev_chunk_sectors);
 
 	/* We update the metadata at least every 10 seconds, or when
 	 * the data about to be copied would over-write the source of
@@ -5366,11 +5368,16 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	safepos = conf->reshape_safe;
 	sector_div(safepos, data_disks);
 	if (mddev->reshape_backwards) {
-		writepos -= min_t(sector_t, reshape_sectors, writepos);
+		BUG_ON(writepos < reshape_sectors);
+		writepos -= reshape_sectors;
 		readpos += reshape_sectors;
 		safepos += reshape_sectors;
 	} else {
 		writepos += reshape_sectors;
+		/* readpos and safepos are worst-case calculations.
+		 * A negative number is overly pessimistic, and causes
+		 * obvious problems for unsigned storage.  So clip to 0.
+		 */
 		readpos -= min_t(sector_t, reshape_sectors, readpos);
 		safepos -= min_t(sector_t, reshape_sectors, safepos);
 	}
@@ -5513,7 +5520,10 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	 * then we need to write out the superblock.
 	 */
 	sector_nr += reshape_sectors;
-	if ((sector_nr - mddev->curr_resync_completed) * 2
+	retn = reshape_sectors;
+finish:
+	if (mddev->curr_resync_completed > mddev->resync_max ||
+	    (sector_nr - mddev->curr_resync_completed) * 2
 	    >= mddev->resync_max - mddev->curr_resync_completed) {
 		/* Cannot proceed until we've updated the superblock... */
 		wait_event(conf->wait_for_overlap,
@@ -5538,7 +5548,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 		sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 	}
 ret:
-	return reshape_sectors;
+	return retn;
 }
 
 static inline sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipped)
@@ -5794,6 +5804,18 @@ static void raid5d(struct md_thread *thread)
 
 	md_check_recovery(mddev);
 
+	if (!bio_list_empty(&conf->return_bi) &&
+	    !test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+		struct bio_list tmp = BIO_EMPTY_LIST;
+		spin_lock_irq(&conf->device_lock);
+		if (!test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+			bio_list_merge(&tmp, &conf->return_bi);
+			bio_list_init(&conf->return_bi);
+		}
+		spin_unlock_irq(&conf->device_lock);
+		return_io(&tmp);
+	}
+
 	blk_start_plug(&plug);
 	handled = 0;
 	spin_lock_irq(&conf->device_lock);
@@ -6234,8 +6256,8 @@ raid5_size(struct mddev *mddev, sector_t sectors, int raid_disks)
 		/* size is defined by the smallest of previous and new size */
 		raid_disks = min(conf->raid_disks, conf->previous_raid_disks);
 
-	sectors &= ~((sector_t)mddev->chunk_sectors - 1);
-	sectors &= ~((sector_t)mddev->new_chunk_sectors - 1);
+	sectors &= ~((sector_t)conf->chunk_sectors - 1);
+	sectors &= ~((sector_t)conf->prev_chunk_sectors - 1);
 	return sectors * (raid_disks - conf->max_degraded);
 }
 
@@ -6453,6 +6475,7 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	INIT_LIST_HEAD(&conf->hold_list);
 	INIT_LIST_HEAD(&conf->delayed_list);
 	INIT_LIST_HEAD(&conf->bitmap_list);
+	bio_list_init(&conf->return_bi);
 	init_llist_head(&conf->released_stripes);
 	atomic_set(&conf->active_stripes, 0);
 	atomic_set(&conf->preread_active_stripes, 0);
@@ -6542,6 +6565,9 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	if (conf->reshape_progress != MaxSector) {
 		conf->prev_chunk_sectors = mddev->chunk_sectors;
 		conf->prev_algo = mddev->layout;
+	} else {
+		conf->prev_chunk_sectors = conf->chunk_sectors;
+		conf->prev_algo = conf->algorithm;
 	}
 
 	conf->min_nr_stripes = NR_STRIPES;
@@ -6661,6 +6687,8 @@ static int run(struct mddev *mddev)
 		sector_t here_new, here_old;
 		int old_disks;
 		int max_degraded = (mddev->level == 6 ? 2 : 1);
+		int chunk_sectors;
+		int new_data_disks;
 
 		if (mddev->new_level != mddev->level) {
 			printk(KERN_ERR "md/raid:%s: unsupported reshape "
@@ -6672,28 +6700,25 @@ static int run(struct mddev *mddev)
 		/* reshape_position must be on a new-stripe boundary, and one
 		 * further up in new geometry must map after here in old
 		 * geometry.
+		 * If the chunk sizes are different, then as we perform reshape
+		 * in units of the largest of the two, reshape_position needs
+		 * be a multiple of the largest chunk size times new data disks.
 		 */
 		here_new = mddev->reshape_position;
-		if (sector_div(here_new, mddev->new_chunk_sectors *
-			       (mddev->raid_disks - max_degraded))) {
+		chunk_sectors = max(mddev->chunk_sectors, mddev->new_chunk_sectors);
+		new_data_disks = mddev->raid_disks - max_degraded;
+		if (sector_div(here_new, chunk_sectors * new_data_disks)) {
 			printk(KERN_ERR "md/raid:%s: reshape_position not "
 			       "on a stripe boundary\n", mdname(mddev));
 			return -EINVAL;
 		}
-		reshape_offset = here_new * mddev->new_chunk_sectors;
+		reshape_offset = here_new * chunk_sectors;
 		/* here_new is the stripe we will write to */
 		here_old = mddev->reshape_position;
-		sector_div(here_old, mddev->chunk_sectors *
-			   (old_disks-max_degraded));
+		sector_div(here_old, chunk_sectors * (old_disks-max_degraded));
 		/* here_old is the first stripe that we might need to read
 		 * from */
 		if (mddev->delta_disks == 0) {
-			if ((here_new * mddev->new_chunk_sectors !=
-			     here_old * mddev->chunk_sectors)) {
-				printk(KERN_ERR "md/raid:%s: reshape position is"
-				       " confused - aborting\n", mdname(mddev));
-				return -EINVAL;
-			}
 			/* We cannot be sure it is safe to start an in-place
 			 * reshape.  It is only safe if user-space is monitoring
 			 * and taking constant backups.
@@ -6712,10 +6737,10 @@ static int run(struct mddev *mddev)
 				return -EINVAL;
 			}
 		} else if (mddev->reshape_backwards
-		    ? (here_new * mddev->new_chunk_sectors + min_offset_diff <=
-		       here_old * mddev->chunk_sectors)
-		    : (here_new * mddev->new_chunk_sectors >=
-		       here_old * mddev->chunk_sectors + (-min_offset_diff))) {
+		    ? (here_new * chunk_sectors + min_offset_diff <=
+		       here_old * chunk_sectors)
+		    : (here_new * chunk_sectors >=
+		       here_old * chunk_sectors + (-min_offset_diff))) {
 			/* Reading from the same stripe as writing to - bad */
 			printk(KERN_ERR "md/raid:%s: reshape_position too early for "
 			       "auto-recovery - aborting.\n",
@@ -6967,7 +6992,7 @@ static void status(struct seq_file *seq, struct mddev *mddev)
 	int i;
 
 	seq_printf(seq, " level %d, %dk chunk, algorithm %d", mddev->level,
-		mddev->chunk_sectors / 2, mddev->layout);
+		conf->chunk_sectors / 2, mddev->layout);
 	seq_printf (seq, " [%d/%d] [", conf->raid_disks, conf->raid_disks - mddev->degraded);
 	for (i = 0; i < conf->raid_disks; i++)
 		seq_printf (seq, "%s",
@@ -7173,7 +7198,9 @@ static int raid5_resize(struct mddev *mddev, sector_t sectors)
 	 * worth it.
 	 */
 	sector_t newsize;
-	sectors &= ~((sector_t)mddev->chunk_sectors - 1);
+	struct r5conf *conf = mddev->private;
+
+	sectors &= ~((sector_t)conf->chunk_sectors - 1);
 	newsize = raid5_size(mddev, sectors, mddev->raid_disks);
 	if (mddev->external_size &&
 	    mddev->array_sectors > newsize)
@@ -7412,6 +7439,7 @@ static void end_reshape(struct r5conf *conf)
 			rdev->data_offset = rdev->new_data_offset;
 		smp_wmb();
 		conf->reshape_progress = MaxSector;
+		conf->mddev->reshape_position = MaxSector;
 		spin_unlock_irq(&conf->device_lock);
 		wake_up(&conf->wait_for_overlap);
 
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index d05144278690..828c2925e68f 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -265,7 +265,7 @@ struct stripe_head_state {
 	int dec_preread_active;
 	unsigned long ops_request;
 
-	struct bio *return_bi;
+	struct bio_list return_bi;
 	struct md_rdev *blocked_rdev;
 	int handle_bad_blocks;
 };
@@ -476,6 +476,9 @@ struct r5conf {
 	int			skip_copy; /* Don't copy data from bio to stripe cache */
 	struct list_head	*last_hold; /* detect hold_list promotions */
 
+	/* bios to have bi_end_io called after metadata is synced */
+	struct bio_list		return_bi;
+
 	atomic_t		reshape_stripes; /* stripes with pending writes for reshape */
 	/* unfortunately we need two cache names as we temporarily have
 	 * two caches.