Merge tag 'drivers-5.10-2020-10-12' of git://git.kernel.dk/linux-block

Pull block driver updates from Jens Axboe:
 "Here are the driver updates for 5.10.

  A few SCSI updates in here too, in coordination with Martin as they
  depend on core block changes for the shared tag bitmap.

  This contains:

   - NVMe pull requests via Christoph:
      - fix keep alive timer modification (Amit Engel)
      - order the PCI ID list more sensibly (Andy Shevchenko)
      - cleanup the open by controller helper (Chaitanya Kulkarni)
      - use an xarray for the CSE log lookup (Chaitanya Kulkarni)
      - support ZNS in nvmet passthrough mode (Chaitanya Kulkarni)
      - fix nvme_ns_report_zones (Christoph Hellwig)
      - add a sanity check to nvmet-fc (James Smart)
      - fix interrupt allocation when too many polled queues are
        specified (Jeffle Xu)
      - small nvmet-tcp optimization (Mark Wunderlich)
      - fix a controller refcount leak on init failure (Chaitanya
        Kulkarni)
      - misc cleanups (Chaitanya Kulkarni)
      - major refactoring of the scanning code (Christoph Hellwig)

   - MD updates via Song:
      - Bug fixes in bitmap code, from Zhao Heming
      - Fix a work queue check, from Guoqing Jiang
      - Fix raid5 oops with reshape, from Song Liu
      - Clean up unused code, from Jason Yan
      - Discard improvements, from Xiao Ni
      - raid5/6 page offset support, from Yufen Yu

   - Shared tag bitmap for SCSI/hisi_sas/null_blk (John, Kashyap,
     Hannes)

   - null_blk open/active zone limit support (Niklas)

   - Set of bcache updates (Coly, Dongsheng, Qinglang)"

* tag 'drivers-5.10-2020-10-12' of git://git.kernel.dk/linux-block: (78 commits)
  md/raid5: fix oops during stripe resizing
  md/bitmap: fix memory leak of temporary bitmap
  md: fix the checking of wrong work queue
  md/bitmap: md_bitmap_get_counter returns wrong blocks
  md/bitmap: md_bitmap_read_sb uses wrong bitmap blocks
  md/raid0: remove unused function is_io_in_chunk_boundary()
  nvme-core: remove extra condition for vwc
  nvme-core: remove extra variable
  nvme: remove nvme_identify_ns_list
  nvme: refactor nvme_validate_ns
  nvme: move nvme_validate_ns
  nvme: query namespace identifiers before adding the namespace
  nvme: revalidate zone bitmaps in nvme_update_ns_info
  nvme: remove nvme_update_formats
  nvme: update the known admin effects
  nvme: set the queue limits in nvme_update_ns_info
  nvme: remove the 0 lba_shift check in nvme_update_ns_info
  nvme: clean up the check for too large logic block sizes
  nvme: freeze the queue over ->lba_shift updates
  nvme: factor out a nvme_configure_metadata helper
  ...

48 files changed, 1766 insertions, 1164 deletions

diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h
index daed4a9c3436..d2e7db43a52a 100644
--- a/drivers/block/null_blk.h
+++ b/drivers/block/null_blk.h
@@ -42,6 +42,9 @@ struct nullb_device {
 	struct badblocks badblocks;
 
 	unsigned int nr_zones;
+	unsigned int nr_zones_imp_open;
+	unsigned int nr_zones_exp_open;
+	unsigned int nr_zones_closed;
 	struct blk_zone *zones;
 	sector_t zone_size_sects;
 
@@ -51,6 +54,8 @@ struct nullb_device {
 	unsigned long zone_size; /* zone size in MB if device is zoned */
 	unsigned long zone_capacity; /* zone capacity in MB if device is zoned */
 	unsigned int zone_nr_conv; /* number of conventional zones */
+	unsigned int zone_max_open; /* max number of open zones */
+	unsigned int zone_max_active; /* max number of active zones */
 	unsigned int submit_queues; /* number of submission queues */
 	unsigned int home_node; /* home node for the device */
 	unsigned int queue_mode; /* block interface */
diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c
index d74443a9c8fa..4685ea401d5b 100644
--- a/drivers/block/null_blk_main.c
+++ b/drivers/block/null_blk_main.c
@@ -164,6 +164,10 @@ static bool shared_tags;
 module_param(shared_tags, bool, 0444);
 MODULE_PARM_DESC(shared_tags, "Share tag set between devices for blk-mq");
 
+static bool g_shared_tag_bitmap;
+module_param_named(shared_tag_bitmap, g_shared_tag_bitmap, bool, 0444);
+MODULE_PARM_DESC(shared_tag_bitmap, "Use shared tag bitmap for all submission queues for blk-mq");
+
 static int g_irqmode = NULL_IRQ_SOFTIRQ;
 
 static int null_set_irqmode(const char *str, const struct kernel_param *kp)
@@ -208,6 +212,14 @@ static unsigned int g_zone_nr_conv;
 module_param_named(zone_nr_conv, g_zone_nr_conv, uint, 0444);
 MODULE_PARM_DESC(zone_nr_conv, "Number of conventional zones when block device is zoned. Default: 0");
 
+static unsigned int g_zone_max_open;
+module_param_named(zone_max_open, g_zone_max_open, uint, 0444);
+MODULE_PARM_DESC(zone_max_open, "Maximum number of open zones when block device is zoned. Default: 0 (no limit)");
+
+static unsigned int g_zone_max_active;
+module_param_named(zone_max_active, g_zone_max_active, uint, 0444);
+MODULE_PARM_DESC(zone_max_active, "Maximum number of active zones when block device is zoned. Default: 0 (no limit)");
+
 static struct nullb_device *null_alloc_dev(void);
 static void null_free_dev(struct nullb_device *dev);
 static void null_del_dev(struct nullb *nullb);
@@ -347,6 +359,8 @@ NULLB_DEVICE_ATTR(zoned, bool, NULL);
 NULLB_DEVICE_ATTR(zone_size, ulong, NULL);
 NULLB_DEVICE_ATTR(zone_capacity, ulong, NULL);
 NULLB_DEVICE_ATTR(zone_nr_conv, uint, NULL);
+NULLB_DEVICE_ATTR(zone_max_open, uint, NULL);
+NULLB_DEVICE_ATTR(zone_max_active, uint, NULL);
 
 static ssize_t nullb_device_power_show(struct config_item *item, char *page)
 {
@@ -464,6 +478,8 @@ static struct configfs_attribute *nullb_device_attrs[] = {
 	&nullb_device_attr_zone_size,
 	&nullb_device_attr_zone_capacity,
 	&nullb_device_attr_zone_nr_conv,
+	&nullb_device_attr_zone_max_open,
+	&nullb_device_attr_zone_max_active,
 	NULL,
 };
 
@@ -517,7 +533,7 @@ nullb_group_drop_item(struct config_group *group, struct config_item *item)
 static ssize_t memb_group_features_show(struct config_item *item, char *page)
 {
 	return snprintf(page, PAGE_SIZE,
-			"memory_backed,discard,bandwidth,cache,badblocks,zoned,zone_size,zone_capacity,zone_nr_conv\n");
+			"memory_backed,discard,bandwidth,cache,badblocks,zoned,zone_size,zone_capacity,zone_nr_conv,zone_max_open,zone_max_active\n");
 }
 
 CONFIGFS_ATTR_RO(memb_group_, features);
@@ -580,6 +596,8 @@ static struct nullb_device *null_alloc_dev(void)
 	dev->zone_size = g_zone_size;
 	dev->zone_capacity = g_zone_capacity;
 	dev->zone_nr_conv = g_zone_nr_conv;
+	dev->zone_max_open = g_zone_max_open;
+	dev->zone_max_active = g_zone_max_active;
 	return dev;
 }
 
@@ -1692,6 +1710,8 @@ static int null_init_tag_set(struct nullb *nullb, struct blk_mq_tag_set *set)
 	set->flags = BLK_MQ_F_SHOULD_MERGE;
 	if (g_no_sched)
 		set->flags |= BLK_MQ_F_NO_SCHED;
+	if (g_shared_tag_bitmap)
+		set->flags |= BLK_MQ_F_TAG_HCTX_SHARED;
 	set->driver_data = NULL;
 
 	if ((nullb && nullb->dev->blocking) || g_blocking)
diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c
index 3d25c9ad2383..fa0cc70f05e6 100644
--- a/drivers/block/null_blk_zoned.c
+++ b/drivers/block/null_blk_zoned.c
@@ -51,6 +51,22 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
 			dev->zone_nr_conv);
 	}
 
+	/* Max active zones has to be < nbr of seq zones in order to be enforceable */
+	if (dev->zone_max_active >= dev->nr_zones - dev->zone_nr_conv) {
+		dev->zone_max_active = 0;
+		pr_info("zone_max_active limit disabled, limit >= zone count\n");
+	}
+
+	/* Max open zones has to be <= max active zones */
+	if (dev->zone_max_active && dev->zone_max_open > dev->zone_max_active) {
+		dev->zone_max_open = dev->zone_max_active;
+		pr_info("changed the maximum number of open zones to %u\n",
+			dev->nr_zones);
+	} else if (dev->zone_max_open >= dev->nr_zones - dev->zone_nr_conv) {
+		dev->zone_max_open = 0;
+		pr_info("zone_max_open limit disabled, limit >= zone count\n");
+	}
+
 	for (i = 0; i <  dev->zone_nr_conv; i++) {
 		struct blk_zone *zone = &dev->zones[i];
 
@@ -99,6 +115,8 @@ int null_register_zoned_dev(struct nullb *nullb)
 	}
 
 	blk_queue_max_zone_append_sectors(q, dev->zone_size_sects);
+	blk_queue_max_open_zones(q, dev->zone_max_open);
+	blk_queue_max_active_zones(q, dev->zone_max_active);
 
 	return 0;
 }
@@ -159,6 +177,103 @@ size_t null_zone_valid_read_len(struct nullb *nullb,
 	return (zone->wp - sector) << SECTOR_SHIFT;
 }
 
+static blk_status_t null_close_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+	if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+		return BLK_STS_IOERR;
+
+	switch (zone->cond) {
+	case BLK_ZONE_COND_CLOSED:
+		/* close operation on closed is not an error */
+		return BLK_STS_OK;
+	case BLK_ZONE_COND_IMP_OPEN:
+		dev->nr_zones_imp_open--;
+		break;
+	case BLK_ZONE_COND_EXP_OPEN:
+		dev->nr_zones_exp_open--;
+		break;
+	case BLK_ZONE_COND_EMPTY:
+	case BLK_ZONE_COND_FULL:
+	default:
+		return BLK_STS_IOERR;
+	}
+
+	if (zone->wp == zone->start) {
+		zone->cond = BLK_ZONE_COND_EMPTY;
+	} else {
+		zone->cond = BLK_ZONE_COND_CLOSED;
+		dev->nr_zones_closed++;
+	}
+
+	return BLK_STS_OK;
+}
+
+static void null_close_first_imp_zone(struct nullb_device *dev)
+{
+	unsigned int i;
+
+	for (i = dev->zone_nr_conv; i < dev->nr_zones; i++) {
+		if (dev->zones[i].cond == BLK_ZONE_COND_IMP_OPEN) {
+			null_close_zone(dev, &dev->zones[i]);
+			return;
+		}
+	}
+}
+
+static bool null_can_set_active(struct nullb_device *dev)
+{
+	if (!dev->zone_max_active)
+		return true;
+
+	return dev->nr_zones_exp_open + dev->nr_zones_imp_open +
+	       dev->nr_zones_closed < dev->zone_max_active;
+}
+
+static bool null_can_open(struct nullb_device *dev)
+{
+	if (!dev->zone_max_open)
+		return true;
+
+	if (dev->nr_zones_exp_open + dev->nr_zones_imp_open < dev->zone_max_open)
+		return true;
+
+	if (dev->nr_zones_imp_open && null_can_set_active(dev)) {
+		null_close_first_imp_zone(dev);
+		return true;
+	}
+
+	return false;
+}
+
+/*
+ * This function matches the manage open zone resources function in the ZBC standard,
+ * with the addition of max active zones support (added in the ZNS standard).
+ *
+ * The function determines if a zone can transition to implicit open or explicit open,
+ * while maintaining the max open zone (and max active zone) limit(s). It may close an
+ * implicit open zone in order to make additional zone resources available.
+ *
+ * ZBC states that an implicit open zone shall be closed only if there is not
+ * room within the open limit. However, with the addition of an active limit,
+ * it is not certain that closing an implicit open zone will allow a new zone
+ * to be opened, since we might already be at the active limit capacity.
+ */
+static bool null_has_zone_resources(struct nullb_device *dev, struct blk_zone *zone)
+{
+	switch (zone->cond) {
+	case BLK_ZONE_COND_EMPTY:
+		if (!null_can_set_active(dev))
+			return false;
+		fallthrough;
+	case BLK_ZONE_COND_CLOSED:
+		return null_can_open(dev);
+	default:
+		/* Should never be called for other states */
+		WARN_ON(1);
+		return false;
+	}
+}
+
 static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
 				    unsigned int nr_sectors, bool append)
 {
@@ -177,43 +292,155 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
 		/* Cannot write to a full zone */
 		return BLK_STS_IOERR;
 	case BLK_ZONE_COND_EMPTY:
+	case BLK_ZONE_COND_CLOSED:
+		if (!null_has_zone_resources(dev, zone))
+			return BLK_STS_IOERR;
+		break;
 	case BLK_ZONE_COND_IMP_OPEN:
 	case BLK_ZONE_COND_EXP_OPEN:
+		break;
+	default:
+		/* Invalid zone condition */
+		return BLK_STS_IOERR;
+	}
+
+	/*
+	 * Regular writes must be at the write pointer position.
+	 * Zone append writes are automatically issued at the write
+	 * pointer and the position returned using the request or BIO
+	 * sector.
+	 */
+	if (append) {
+		sector = zone->wp;
+		if (cmd->bio)
+			cmd->bio->bi_iter.bi_sector = sector;
+		else
+			cmd->rq->__sector = sector;
+	} else if (sector != zone->wp) {
+		return BLK_STS_IOERR;
+	}
+
+	if (zone->wp + nr_sectors > zone->start + zone->capacity)
+		return BLK_STS_IOERR;
+
+	if (zone->cond == BLK_ZONE_COND_CLOSED) {
+		dev->nr_zones_closed--;
+		dev->nr_zones_imp_open++;
+	} else if (zone->cond == BLK_ZONE_COND_EMPTY) {
+		dev->nr_zones_imp_open++;
+	}
+	if (zone->cond != BLK_ZONE_COND_EXP_OPEN)
+		zone->cond = BLK_ZONE_COND_IMP_OPEN;
+
+	ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
+	if (ret != BLK_STS_OK)
+		return ret;
+
+	zone->wp += nr_sectors;
+	if (zone->wp == zone->start + zone->capacity) {
+		if (zone->cond == BLK_ZONE_COND_EXP_OPEN)
+			dev->nr_zones_exp_open--;
+		else if (zone->cond == BLK_ZONE_COND_IMP_OPEN)
+			dev->nr_zones_imp_open--;
+		zone->cond = BLK_ZONE_COND_FULL;
+	}
+	return BLK_STS_OK;
+}
+
+static blk_status_t null_open_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+	if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+		return BLK_STS_IOERR;
+
+	switch (zone->cond) {
+	case BLK_ZONE_COND_EXP_OPEN:
+		/* open operation on exp open is not an error */
+		return BLK_STS_OK;
+	case BLK_ZONE_COND_EMPTY:
+		if (!null_has_zone_resources(dev, zone))
+			return BLK_STS_IOERR;
+		break;
+	case BLK_ZONE_COND_IMP_OPEN:
+		dev->nr_zones_imp_open--;
+		break;
 	case BLK_ZONE_COND_CLOSED:
-		/*
-		 * Regular writes must be at the write pointer position.
-		 * Zone append writes are automatically issued at the write
-		 * pointer and the position returned using the request or BIO
-		 * sector.
-		 */
-		if (append) {
-			sector = zone->wp;
-			if (cmd->bio)
-				cmd->bio->bi_iter.bi_sector = sector;
-			else
-				cmd->rq->__sector = sector;
-		} else if (sector != zone->wp) {
+		if (!null_has_zone_resources(dev, zone))
 			return BLK_STS_IOERR;
-		}
+		dev->nr_zones_closed--;
+		break;
+	case BLK_ZONE_COND_FULL:
+	default:
+		return BLK_STS_IOERR;
+	}
+
+	zone->cond = BLK_ZONE_COND_EXP_OPEN;
+	dev->nr_zones_exp_open++;
 
-		if (zone->wp + nr_sectors > zone->start + zone->capacity)
+	return BLK_STS_OK;
+}
+
+static blk_status_t null_finish_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+	if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+		return BLK_STS_IOERR;
+
+	switch (zone->cond) {
+	case BLK_ZONE_COND_FULL:
+		/* finish operation on full is not an error */
+		return BLK_STS_OK;
+	case BLK_ZONE_COND_EMPTY:
+		if (!null_has_zone_resources(dev, zone))
 			return BLK_STS_IOERR;
+		break;
+	case BLK_ZONE_COND_IMP_OPEN:
+		dev->nr_zones_imp_open--;
+		break;
+	case BLK_ZONE_COND_EXP_OPEN:
+		dev->nr_zones_exp_open--;
+		break;
+	case BLK_ZONE_COND_CLOSED:
+		if (!null_has_zone_resources(dev, zone))
+			return BLK_STS_IOERR;
+		dev->nr_zones_closed--;
+		break;
+	default:
+		return BLK_STS_IOERR;
+	}
 
-		if (zone->cond != BLK_ZONE_COND_EXP_OPEN)
-			zone->cond = BLK_ZONE_COND_IMP_OPEN;
+	zone->cond = BLK_ZONE_COND_FULL;
+	zone->wp = zone->start + zone->len;
 
-		ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
-		if (ret != BLK_STS_OK)
-			return ret;
+	return BLK_STS_OK;
+}
+
+static blk_status_t null_reset_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+	if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+		return BLK_STS_IOERR;
 
-		zone->wp += nr_sectors;
-		if (zone->wp == zone->start + zone->capacity)
-			zone->cond = BLK_ZONE_COND_FULL;
+	switch (zone->cond) {
+	case BLK_ZONE_COND_EMPTY:
+		/* reset operation on empty is not an error */
 		return BLK_STS_OK;
+	case BLK_ZONE_COND_IMP_OPEN:
+		dev->nr_zones_imp_open--;
+		break;
+	case BLK_ZONE_COND_EXP_OPEN:
+		dev->nr_zones_exp_open--;
+		break;
+	case BLK_ZONE_COND_CLOSED:
+		dev->nr_zones_closed--;
+		break;
+	case BLK_ZONE_COND_FULL:
+		break;
 	default:
-		/* Invalid zone condition */
 		return BLK_STS_IOERR;
 	}
+
+	zone->cond = BLK_ZONE_COND_EMPTY;
+	zone->wp = zone->start;
+
+	return BLK_STS_OK;
 }
 
 static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
@@ -222,56 +449,34 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
 	struct nullb_device *dev = cmd->nq->dev;
 	unsigned int zone_no = null_zone_no(dev, sector);
 	struct blk_zone *zone = &dev->zones[zone_no];
+	blk_status_t ret = BLK_STS_OK;
 	size_t i;
 
 	switch (op) {
 	case REQ_OP_ZONE_RESET_ALL:
-		for (i = 0; i < dev->nr_zones; i++) {
-			if (zone[i].type == BLK_ZONE_TYPE_CONVENTIONAL)
-				continue;
-			zone[i].cond = BLK_ZONE_COND_EMPTY;
-			zone[i].wp = zone[i].start;
-		}
+		for (i = dev->zone_nr_conv; i < dev->nr_zones; i++)
+			null_reset_zone(dev, &dev->zones[i]);
 		break;
 	case REQ_OP_ZONE_RESET:
-		if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
-			return BLK_STS_IOERR;
-
-		zone->cond = BLK_ZONE_COND_EMPTY;
-		zone->wp = zone->start;
+		ret = null_reset_zone(dev, zone);
 		break;
 	case REQ_OP_ZONE_OPEN:
-		if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
-			return BLK_STS_IOERR;
-		if (zone->cond == BLK_ZONE_COND_FULL)
-			return BLK_STS_IOERR;
-
-		zone->cond = BLK_ZONE_COND_EXP_OPEN;
+		ret = null_open_zone(dev, zone);
 		break;
 	case REQ_OP_ZONE_CLOSE:
-		if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
-			return BLK_STS_IOERR;
-		if (zone->cond == BLK_ZONE_COND_FULL)
-			return BLK_STS_IOERR;
-
-		if (zone->wp == zone->start)
-			zone->cond = BLK_ZONE_COND_EMPTY;
-		else
-			zone->cond = BLK_ZONE_COND_CLOSED;
+		ret = null_close_zone(dev, zone);
 		break;
 	case REQ_OP_ZONE_FINISH:
-		if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
-			return BLK_STS_IOERR;
-
-		zone->cond = BLK_ZONE_COND_FULL;
-		zone->wp = zone->start + zone->len;
+		ret = null_finish_zone(dev, zone);
 		break;
 	default:
 		return BLK_STS_NOTSUPP;
 	}
 
-	trace_nullb_zone_op(cmd, zone_no, zone->cond);
-	return BLK_STS_OK;
+	if (ret == BLK_STS_OK)
+		trace_nullb_zone_op(cmd, zone_no, zone->cond);
+
+	return ret;
 }
 
 blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op,
diff --git a/drivers/block/rsxx/core.c b/drivers/block/rsxx/core.c
index 8799e3bab067..63f549889f87 100644
--- a/drivers/block/rsxx/core.c
+++ b/drivers/block/rsxx/core.c
@@ -439,7 +439,7 @@ static void card_state_change(struct rsxx_cardinfo *card,
 	case CARD_STATE_FAULT:
 		dev_crit(CARD_TO_DEV(card),
 			"Hardware Fault reported!\n");
-		/* Fall through. */
+		fallthrough;
 
 	/* Everything else, detach DMA interface if it's attached. */
 	case CARD_STATE_SHUTDOWN:
diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index 52035a78d836..8c371d5eef8e 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -49,7 +49,7 @@
  *
  * bch_bucket_alloc() allocates a single bucket from a specific cache.
  *
- * bch_bucket_alloc_set() allocates one or more buckets from different caches
+ * bch_bucket_alloc_set() allocates one  bucket from different caches
  * out of a cache set.
  *
  * free_some_buckets() drives all the processes described above. It's called
@@ -87,8 +87,7 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
 {
 	struct cache *ca;
 	struct bucket *b;
-	unsigned long next = c->nbuckets * c->sb.bucket_size / 1024;
-	unsigned int i;
+	unsigned long next = c->nbuckets * c->cache->sb.bucket_size / 1024;
 	int r;
 
 	atomic_sub(sectors, &c->rescale);
@@ -104,14 +103,14 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
 
 	c->min_prio = USHRT_MAX;
 
-	for_each_cache(ca, c, i)
-		for_each_bucket(b, ca)
-			if (b->prio &&
-			    b->prio != BTREE_PRIO &&
-			    !atomic_read(&b->pin)) {
-				b->prio--;
-				c->min_prio = min(c->min_prio, b->prio);
-			}
+	ca = c->cache;
+	for_each_bucket(b, ca)
+		if (b->prio &&
+		    b->prio != BTREE_PRIO &&
+		    !atomic_read(&b->pin)) {
+			b->prio--;
+			c->min_prio = min(c->min_prio, b->prio);
+		}
 
 	mutex_unlock(&c->bucket_lock);
 }
@@ -362,7 +361,7 @@ retry_invalidate:
 		 * new stuff to them:
 		 */
 		allocator_wait(ca, !atomic_read(&ca->set->prio_blocked));
-		if (CACHE_SYNC(&ca->set->sb)) {
+		if (CACHE_SYNC(&ca->sb)) {
 			/*
 			 * This could deadlock if an allocation with a btree
 			 * node locked ever blocked - having the btree node
@@ -488,34 +487,29 @@ void bch_bucket_free(struct cache_set *c, struct bkey *k)
 }
 
 int __bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			   struct bkey *k, int n, bool wait)
+			   struct bkey *k, bool wait)
 {
-	int i;
+	struct cache *ca;
+	long b;
 
 	/* No allocation if CACHE_SET_IO_DISABLE bit is set */
 	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags)))
 		return -1;
 
 	lockdep_assert_held(&c->bucket_lock);
-	BUG_ON(!n || n > c->caches_loaded || n > MAX_CACHES_PER_SET);
 
 	bkey_init(k);
 
-	/* sort by free space/prio of oldest data in caches */
-
-	for (i = 0; i < n; i++) {
-		struct cache *ca = c->cache_by_alloc[i];
-		long b = bch_bucket_alloc(ca, reserve, wait);
+	ca = c->cache;
+	b = bch_bucket_alloc(ca, reserve, wait);
+	if (b == -1)
+		goto err;
 
-		if (b == -1)
-			goto err;
+	k->ptr[0] = MAKE_PTR(ca->buckets[b].gen,
+			     bucket_to_sector(c, b),
+			     ca->sb.nr_this_dev);
 
-		k->ptr[i] = MAKE_PTR(ca->buckets[b].gen,
-				bucket_to_sector(c, b),
-				ca->sb.nr_this_dev);
-
-		SET_KEY_PTRS(k, i + 1);
-	}
+	SET_KEY_PTRS(k, 1);
 
 	return 0;
 err:
@@ -525,12 +519,12 @@ err:
 }
 
 int bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			 struct bkey *k, int n, bool wait)
+			 struct bkey *k, bool wait)
 {
 	int ret;
 
 	mutex_lock(&c->bucket_lock);
-	ret = __bch_bucket_alloc_set(c, reserve, k, n, wait);
+	ret = __bch_bucket_alloc_set(c, reserve, k, wait);
 	mutex_unlock(&c->bucket_lock);
 	return ret;
 }
@@ -589,7 +583,7 @@ static struct open_bucket *pick_data_bucket(struct cache_set *c,
 					   struct open_bucket, list);
 found:
 	if (!ret->sectors_free && KEY_PTRS(alloc)) {
-		ret->sectors_free = c->sb.bucket_size;
+		ret->sectors_free = c->cache->sb.bucket_size;
 		bkey_copy(&ret->key, alloc);
 		bkey_init(alloc);
 	}
@@ -638,7 +632,7 @@ bool bch_alloc_sectors(struct cache_set *c,
 
 		spin_unlock(&c->data_bucket_lock);
 
-		if (bch_bucket_alloc_set(c, watermark, &alloc.key, 1, wait))
+		if (bch_bucket_alloc_set(c, watermark, &alloc.key, wait))
 			return false;
 
 		spin_lock(&c->data_bucket_lock);
@@ -683,7 +677,7 @@ bool bch_alloc_sectors(struct cache_set *c,
 				&PTR_CACHE(c, &b->key, i)->sectors_written);
 	}
 
-	if (b->sectors_free < c->sb.block_size)
+	if (b->sectors_free < c->cache->sb.block_size)
 		b->sectors_free = 0;
 
 	/*
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 4fd03d2496d8..1d57f48307e6 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -517,11 +517,7 @@ struct cache_set {
 	atomic_t		idle_counter;
 	atomic_t		at_max_writeback_rate;
 
-	struct cache_sb		sb;
-
-	struct cache		*cache[MAX_CACHES_PER_SET];
-	struct cache		*cache_by_alloc[MAX_CACHES_PER_SET];
-	int			caches_loaded;
+	struct cache		*cache;
 
 	struct bcache_device	**devices;
 	unsigned int		devices_max_used;
@@ -670,6 +666,7 @@ struct cache_set {
 	struct mutex		verify_lock;
 #endif
 
+	uint8_t			set_uuid[16];
 	unsigned int		nr_uuids;
 	struct uuid_entry	*uuids;
 	BKEY_PADDED(uuid_bucket);
@@ -758,9 +755,8 @@ struct bbio {
 #define btree_default_blocks(c)						\
 	((unsigned int) ((PAGE_SECTORS * (c)->btree_pages) >> (c)->block_bits))
 
-#define bucket_pages(c)		((c)->sb.bucket_size / PAGE_SECTORS)
-#define bucket_bytes(c)		((c)->sb.bucket_size << 9)
-#define block_bytes(c)		((c)->sb.block_size << 9)
+#define bucket_bytes(ca)	((ca)->sb.bucket_size << 9)
+#define block_bytes(ca)		((ca)->sb.block_size << 9)
 
 static inline unsigned int meta_bucket_pages(struct cache_sb *sb)
 {
@@ -801,14 +797,14 @@ static inline sector_t bucket_to_sector(struct cache_set *c, size_t b)
 
 static inline sector_t bucket_remainder(struct cache_set *c, sector_t s)
 {
-	return s & (c->sb.bucket_size - 1);
+	return s & (c->cache->sb.bucket_size - 1);
 }
 
 static inline struct cache *PTR_CACHE(struct cache_set *c,
 				      const struct bkey *k,
 				      unsigned int ptr)
 {
-	return c->cache[PTR_DEV(k, ptr)];
+	return c->cache;
 }
 
 static inline size_t PTR_BUCKET_NR(struct cache_set *c,
@@ -889,9 +885,6 @@ do {									\
 
 /* Looping macros */
 
-#define for_each_cache(ca, cs, iter)					\
-	for (iter = 0; ca = cs->cache[iter], iter < (cs)->sb.nr_in_set; iter++)
-
 #define for_each_bucket(b, ca)						\
 	for (b = (ca)->buckets + (ca)->sb.first_bucket;			\
 	     b < (ca)->buckets + (ca)->sb.nbuckets; b++)
@@ -933,11 +926,9 @@ static inline uint8_t bucket_gc_gen(struct bucket *b)
 
 static inline void wake_up_allocators(struct cache_set *c)
 {
-	struct cache *ca;
-	unsigned int i;
+	struct cache *ca = c->cache;
 
-	for_each_cache(ca, c, i)
-		wake_up_process(ca->alloc_thread);
+	wake_up_process(ca->alloc_thread);
 }
 
 static inline void closure_bio_submit(struct cache_set *c,
@@ -994,9 +985,9 @@ void bch_bucket_free(struct cache_set *c, struct bkey *k);
 
 long bch_bucket_alloc(struct cache *ca, unsigned int reserve, bool wait);
 int __bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			   struct bkey *k, int n, bool wait);
+			   struct bkey *k, bool wait);
 int bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
-			 struct bkey *k, int n, bool wait);
+			 struct bkey *k, bool wait);
 bool bch_alloc_sectors(struct cache_set *c, struct bkey *k,
 		       unsigned int sectors, unsigned int write_point,
 		       unsigned int write_prio, bool wait);
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 3d8bd0692af3..910df242c83d 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -104,7 +104,7 @@
 
 static inline struct bset *write_block(struct btree *b)
 {
-	return ((void *) btree_bset_first(b)) + b->written * block_bytes(b->c);
+	return ((void *) btree_bset_first(b)) + b->written * block_bytes(b->c->cache);
 }
 
 static void bch_btree_init_next(struct btree *b)
@@ -117,7 +117,7 @@ static void bch_btree_init_next(struct btree *b)
 
 	if (b->written < btree_blocks(b))
 		bch_bset_init_next(&b->keys, write_block(b),
-				   bset_magic(&b->c->sb));
+				   bset_magic(&b->c->cache->sb));
 
 }
 
@@ -155,7 +155,7 @@ void bch_btree_node_read_done(struct btree *b)
 	 * See the comment arount cache_set->fill_iter.
 	 */
 	iter = mempool_alloc(&b->c->fill_iter, GFP_NOIO);
-	iter->size = b->c->sb.bucket_size / b->c->sb.block_size;
+	iter->size = b->c->cache->sb.bucket_size / b->c->cache->sb.block_size;
 	iter->used = 0;
 
 #ifdef CONFIG_BCACHE_DEBUG
@@ -173,12 +173,12 @@ void bch_btree_node_read_done(struct btree *b)
 			goto err;
 
 		err = "bad btree header";
-		if (b->written + set_blocks(i, block_bytes(b->c)) >
+		if (b->written + set_blocks(i, block_bytes(b->c->cache)) >
 		    btree_blocks(b))
 			goto err;
 
 		err = "bad magic";
-		if (i->magic != bset_magic(&b->c->sb))
+		if (i->magic != bset_magic(&b->c->cache->sb))
 			goto err;
 
 		err = "bad checksum";
@@ -199,13 +199,13 @@ void bch_btree_node_read_done(struct btree *b)
 
 		bch_btree_iter_push(iter, i->start, bset_bkey_last(i));
 
-		b->written += set_blocks(i, block_bytes(b->c));
+		b->written += set_blocks(i, block_bytes(b->c->cache));
 	}
 
 	err = "corrupted btree";
 	for (i = write_block(b);
 	     bset_sector_offset(&b->keys, i) < KEY_SIZE(&b->key);
-	     i = ((void *) i) + block_bytes(b->c))
+	     i = ((void *) i) + block_bytes(b->c->cache))
 		if (i->seq == b->keys.set[0].data->seq)
 			goto err;
 
@@ -219,7 +219,7 @@ void bch_btree_node_read_done(struct btree *b)
 
 	if (b->written < btree_blocks(b))
 		bch_bset_init_next(&b->keys, write_block(b),
-				   bset_magic(&b->c->sb));
+				   bset_magic(&b->c->cache->sb));
 out:
 	mempool_free(iter, &b->c->fill_iter);
 	return;
@@ -347,7 +347,7 @@ static void do_btree_node_write(struct btree *b)
 
 	b->bio->bi_end_io	= btree_node_write_endio;
 	b->bio->bi_private	= cl;
-	b->bio->bi_iter.bi_size	= roundup(set_bytes(i), block_bytes(b->c));
+	b->bio->bi_iter.bi_size	= roundup(set_bytes(i), block_bytes(b->c->cache));
 	b->bio->bi_opf		= REQ_OP_WRITE | REQ_META | REQ_FUA;
 	bch_bio_map(b->bio, i);
 
@@ -423,10 +423,10 @@ void __bch_btree_node_write(struct btree *b, struct closure *parent)
 
 	do_btree_node_write(b);
 
-	atomic_long_add(set_blocks(i, block_bytes(b->c)) * b->c->sb.block_size,
+	atomic_long_add(set_blocks(i, block_bytes(b->c->cache)) * b->c->cache->sb.block_size,
 			&PTR_CACHE(b->c, &b->key, 0)->btree_sectors_written);
 
-	b->written += set_blocks(i, block_bytes(b->c));
+	b->written += set_blocks(i, block_bytes(b->c->cache));
 }
 
 void bch_btree_node_write(struct btree *b, struct closure *parent)
@@ -514,7 +514,7 @@ static void bch_btree_leaf_dirty(struct btree *b, atomic_t *journal_ref)
  * mca -> memory cache
  */
 
-#define mca_reserve(c)	(((c->root && c->root->level)		\
+#define mca_reserve(c)	(((!IS_ERR_OR_NULL(c->root) && c->root->level) \
 			  ? c->root->level : 1) * 8 + 16)
 #define mca_can_free(c)						\
 	max_t(int, 0, c->btree_cache_used - mca_reserve(c))
@@ -738,7 +738,7 @@ void bch_btree_cache_free(struct cache_set *c)
 	if (c->verify_data)
 		list_move(&c->verify_data->list, &c->btree_cache);
 
-	free_pages((unsigned long) c->verify_ondisk, ilog2(meta_bucket_pages(&c->sb)));
+	free_pages((unsigned long) c->verify_ondisk, ilog2(meta_bucket_pages(&c->cache->sb)));
 #endif
 
 	list_splice(&c->btree_cache_freeable,
@@ -785,7 +785,8 @@ int bch_btree_cache_alloc(struct cache_set *c)
 	mutex_init(&c->verify_lock);
 
 	c->verify_ondisk = (void *)
-		__get_free_pages(GFP_KERNEL|__GFP_COMP, ilog2(meta_bucket_pages(&c->sb)));
+		__get_free_pages(GFP_KERNEL|__GFP_COMP,
+				 ilog2(meta_bucket_pages(&c->cache->sb)));
 	if (!c->verify_ondisk) {
 		/*
 		 * Don't worry about the mca_rereserve buckets
@@ -1091,7 +1092,7 @@ struct btree *__bch_btree_node_alloc(struct cache_set *c, struct btree_op *op,
 
 	mutex_lock(&c->bucket_lock);
 retry:
-	if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, 1, wait))
+	if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, wait))
 		goto err;
 
 	bkey_put(c, &k.key);
@@ -1108,7 +1109,7 @@ retry:
 	}
 
 	b->parent = parent;
-	bch_bset_init_next(&b->keys, b->keys.set->data, bset_magic(&b->c->sb));
+	bch_bset_init_next(&b->keys, b->keys.set->data, bset_magic(&b->c->cache->sb));
 
 	mutex_unlock(&c->bucket_lock);
 
@@ -1167,19 +1168,18 @@ static void make_btree_freeing_key(struct btree *b, struct bkey *k)
 static int btree_check_reserve(struct btree *b, struct btree_op *op)
 {
 	struct cache_set *c = b->c;
-	struct cache *ca;
-	unsigned int i, reserve = (c->root->level - b->level) * 2 + 1;
+	struct cache *ca = c->cache;
+	unsigned int reserve = (c->root->level - b->level) * 2 + 1;
 
 	mutex_lock(&c->bucket_lock);
 
-	for_each_cache(ca, c, i)
-		if (fifo_used(&ca->free[RESERVE_BTREE]) < reserve) {
-			if (op)
-				prepare_to_wait(&c->btree_cache_wait, &op->wait,
-						TASK_UNINTERRUPTIBLE);
-			mutex_unlock(&c->bucket_lock);
-			return -EINTR;
-		}
+	if (fifo_used(&ca->free[RESERVE_BTREE]) < reserve) {
+		if (op)
+			prepare_to_wait(&c->btree_cache_wait, &op->wait,
+					TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&c->bucket_lock);
+		return -EINTR;
+	}
 
 	mutex_unlock(&c->bucket_lock);
 
@@ -1345,7 +1345,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 
 	if (nodes < 2 ||
 	    __set_blocks(b->keys.set[0].data, keys,
-			 block_bytes(b->c)) > blocks * (nodes - 1))
+			 block_bytes(b->c->cache)) > blocks * (nodes - 1))
 		return 0;
 
 	for (i = 0; i < nodes; i++) {
@@ -1379,7 +1379,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			     k = bkey_next(k)) {
 				if (__set_blocks(n1, n1->keys + keys +
 						 bkey_u64s(k),
-						 block_bytes(b->c)) > blocks)
+						 block_bytes(b->c->cache)) > blocks)
 					break;
 
 				last = k;
@@ -1395,7 +1395,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			 * though)
 			 */
 			if (__set_blocks(n1, n1->keys + n2->keys,
-					 block_bytes(b->c)) >
+					 block_bytes(b->c->cache)) >
 			    btree_blocks(new_nodes[i]))
 				goto out_unlock_nocoalesce;
 
@@ -1404,7 +1404,7 @@ static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			last = &r->b->key;
 		}
 
-		BUG_ON(__set_blocks(n1, n1->keys + keys, block_bytes(b->c)) >
+		BUG_ON(__set_blocks(n1, n1->keys + keys, block_bytes(b->c->cache)) >
 		       btree_blocks(new_nodes[i]));
 
 		if (last)
@@ -1695,7 +1695,6 @@ static void btree_gc_start(struct cache_set *c)
 {
 	struct cache *ca;
 	struct bucket *b;
-	unsigned int i;
 
 	if (!c->gc_mark_valid)
 		return;
@@ -1705,14 +1704,14 @@ static void btree_gc_start(struct cache_set *c)
 	c->gc_mark_valid = 0;
 	c->gc_done = ZERO_KEY;
 
-	for_each_cache(ca, c, i)
-		for_each_bucket(b, ca) {
-			b->last_gc = b->gen;
-			if (!atomic_read(&b->pin)) {
-				SET_GC_MARK(b, 0);
-				SET_GC_SECTORS_USED(b, 0);
-			}
+	ca = c->cache;
+	for_each_bucket(b, ca) {
+		b->last_gc = b->gen;
+		if (!atomic_read(&b->pin)) {
+			SET_GC_MARK(b, 0);
+			SET_GC_SECTORS_USED(b, 0);
 		}
+	}
 
 	mutex_unlock(&c->bucket_lock);
 }
@@ -1721,7 +1720,8 @@ static void bch_btree_gc_finish(struct cache_set *c)
 {
 	struct bucket *b;
 	struct cache *ca;
-	unsigned int i;
+	unsigned int i, j;
+	uint64_t *k;
 
 	mutex_lock(&c->bucket_lock);
 
@@ -1739,7 +1739,6 @@ static void bch_btree_gc_finish(struct cache_set *c)
 		struct bcache_device *d = c->devices[i];
 		struct cached_dev *dc;
 		struct keybuf_key *w, *n;
-		unsigned int j;
 
 		if (!d || UUID_FLASH_ONLY(&c->uuids[i]))
 			continue;
@@ -1756,29 +1755,27 @@ static void bch_btree_gc_finish(struct cache_set *c)
 	rcu_read_unlock();
 
 	c->avail_nbuckets = 0;
-	for_each_cache(ca, c, i) {
-		uint64_t *i;
 
-		ca->invalidate_needs_gc = 0;
+	ca = c->cache;
+	ca->invalidate_needs_gc = 0;
 
-		for (i = ca->sb.d; i < ca->sb.d + ca->sb.keys; i++)
-			SET_GC_MARK(ca->buckets + *i, GC_MARK_METADATA);
+	for (k = ca->sb.d; k < ca->sb.d + ca->sb.keys; k++)
+		SET_GC_MARK(ca->buckets + *k, GC_MARK_METADATA);
 
-		for (i = ca->prio_buckets;
-		     i < ca->prio_buckets + prio_buckets(ca) * 2; i++)
-			SET_GC_MARK(ca->buckets + *i, GC_MARK_METADATA);
+	for (k = ca->prio_buckets;
+	     k < ca->prio_buckets + prio_buckets(ca) * 2; k++)
+		SET_GC_MARK(ca->buckets + *k, GC_MARK_METADATA);
 
-		for_each_bucket(b, ca) {
-			c->need_gc	= max(c->need_gc, bucket_gc_gen(b));
+	for_each_bucket(b, ca) {
+		c->need_gc	= max(c->need_gc, bucket_gc_gen(b));
 
-			if (atomic_read(&b->pin))
-				continue;
+		if (atomic_read(&b->pin))
+			continue;
 
-			BUG_ON(!GC_MARK(b) && GC_SECTORS_USED(b));
+		BUG_ON(!GC_MARK(b) && GC_SECTORS_USED(b));
 
-			if (!GC_MARK(b) || GC_MARK(b) == GC_MARK_RECLAIMABLE)
-				c->avail_nbuckets++;
-		}
+		if (!GC_MARK(b) || GC_MARK(b) == GC_MARK_RECLAIMABLE)
+			c->avail_nbuckets++;
 	}
 
 	mutex_unlock(&c->bucket_lock);
@@ -1830,12 +1827,10 @@ static void bch_btree_gc(struct cache_set *c)
 
 static bool gc_should_run(struct cache_set *c)
 {
-	struct cache *ca;
-	unsigned int i;
+	struct cache *ca = c->cache;
 
-	for_each_cache(ca, c, i)
-		if (ca->invalidate_needs_gc)
-			return true;
+	if (ca->invalidate_needs_gc)
+		return true;
 
 	if (atomic_read(&c->sectors_to_gc) < 0)
 		return true;
@@ -2081,9 +2076,8 @@ out:
 
 void bch_initial_gc_finish(struct cache_set *c)
 {
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	struct bucket *b;
-	unsigned int i;
 
 	bch_btree_gc_finish(c);
 
@@ -2098,20 +2092,18 @@ void bch_initial_gc_finish(struct cache_set *c)
 	 * This is only safe for buckets that have no live data in them, which
 	 * there should always be some of.
 	 */
-	for_each_cache(ca, c, i) {
-		for_each_bucket(b, ca) {
-			if (fifo_full(&ca->free[RESERVE_PRIO]) &&
-			    fifo_full(&ca->free[RESERVE_BTREE]))
-				break;
+	for_each_bucket(b, ca) {
+		if (fifo_full(&ca->free[RESERVE_PRIO]) &&
+		    fifo_full(&ca->free[RESERVE_BTREE]))
+			break;
 
-			if (bch_can_invalidate_bucket(ca, b) &&
-			    !GC_MARK(b)) {
-				__bch_invalidate_one_bucket(ca, b);
-				if (!fifo_push(&ca->free[RESERVE_PRIO],
-				   b - ca->buckets))
-					fifo_push(&ca->free[RESERVE_BTREE],
-						  b - ca->buckets);
-			}
+		if (bch_can_invalidate_bucket(ca, b) &&
+		    !GC_MARK(b)) {
+			__bch_invalidate_one_bucket(ca, b);
+			if (!fifo_push(&ca->free[RESERVE_PRIO],
+			   b - ca->buckets))
+				fifo_push(&ca->free[RESERVE_BTREE],
+					  b - ca->buckets);
 		}
 	}
 
@@ -2219,7 +2211,7 @@ static int btree_split(struct btree *b, struct btree_op *op,
 		goto err;
 
 	split = set_blocks(btree_bset_first(n1),
-			   block_bytes(n1->c)) > (btree_blocks(b) * 4) / 5;
+			   block_bytes(n1->c->cache)) > (btree_blocks(b) * 4) / 5;
 
 	if (split) {
 		unsigned int keys = 0;
diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h
index 257969980c49..50482107134f 100644
--- a/drivers/md/bcache/btree.h
+++ b/drivers/md/bcache/btree.h
@@ -194,7 +194,7 @@ static inline unsigned int bset_block_offset(struct btree *b, struct bset *i)
 
 static inline void set_gc_sectors(struct cache_set *c)
 {
-	atomic_set(&c->sectors_to_gc, c->sb.bucket_size * c->nbuckets / 16);
+	atomic_set(&c->sectors_to_gc, c->cache->sb.bucket_size * c->nbuckets / 16);
 }
 
 void bkey_put(struct cache_set *c, struct bkey *k);
diff --git a/drivers/md/bcache/closure.c b/drivers/md/bcache/closure.c
index 0164a1fe94a9..d8d9394a6beb 100644
--- a/drivers/md/bcache/closure.c
+++ b/drivers/md/bcache/closure.c
@@ -159,7 +159,7 @@ void closure_debug_destroy(struct closure *cl)
 
 static struct dentry *closure_debug;
 
-static int debug_seq_show(struct seq_file *f, void *data)
+static int debug_show(struct seq_file *f, void *data)
 {
 	struct closure *cl;
 
@@ -188,17 +188,7 @@ static int debug_seq_show(struct seq_file *f, void *data)
 	return 0;
 }
 
-static int debug_seq_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, debug_seq_show, NULL);
-}
-
-static const struct file_operations debug_ops = {
-	.owner		= THIS_MODULE,
-	.open		= debug_seq_open,
-	.read		= seq_read,
-	.release	= single_release
-};
+DEFINE_SHOW_ATTRIBUTE(debug);
 
 void  __init closure_debug_init(void)
 {
@@ -209,7 +199,7 @@ void  __init closure_debug_init(void)
 		 * about this.
 		 */
 		closure_debug = debugfs_create_file(
-			"closures", 0400, bcache_debug, NULL, &debug_ops);
+			"closures", 0400, bcache_debug, NULL, &debug_fops);
 }
 #endif
 
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index 336f43910383..b00fd08d696b 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -25,8 +25,8 @@ struct dentry *bcache_debug;
 	for (i = (start);						\
 	     (void *) i < (void *) (start) + (KEY_SIZE(&b->key) << 9) &&\
 	     i->seq == (start)->seq;					\
-	     i = (void *) i + set_blocks(i, block_bytes(b->c)) *	\
-		 block_bytes(b->c))
+	     i = (void *) i + set_blocks(i, block_bytes(b->c->cache)) *	\
+		 block_bytes(b->c->cache))
 
 void bch_btree_verify(struct btree *b)
 {
@@ -82,14 +82,14 @@ void bch_btree_verify(struct btree *b)
 
 		for_each_written_bset(b, ondisk, i) {
 			unsigned int block = ((void *) i - (void *) ondisk) /
-				block_bytes(b->c);
+				block_bytes(b->c->cache);
 
 			pr_err("*** on disk block %u:\n", block);
 			bch_dump_bset(&b->keys, i, block);
 		}
 
 		pr_err("*** block %zu not written\n",
-		       ((void *) i - (void *) ondisk) / block_bytes(b->c));
+		       ((void *) i - (void *) ondisk) / block_bytes(b->c->cache));
 
 		for (j = 0; j < inmemory->keys; j++)
 			if (inmemory->d[j] != sorted->d[j])
@@ -238,7 +238,7 @@ void bch_debug_init_cache_set(struct cache_set *c)
 	if (!IS_ERR_OR_NULL(bcache_debug)) {
 		char name[50];
 
-		snprintf(name, 50, "bcache-%pU", c->sb.set_uuid);
+		snprintf(name, 50, "bcache-%pU", c->set_uuid);
 		c->debug = debugfs_create_file(name, 0400, bcache_debug, c,
 					       &cache_set_debug_ops);
 	}
diff --git a/drivers/md/bcache/extents.c b/drivers/md/bcache/extents.c
index 9162af5bb6ec..f4658a1f37b8 100644
--- a/drivers/md/bcache/extents.c
+++ b/drivers/md/bcache/extents.c
@@ -54,7 +54,7 @@ static bool __ptr_invalid(struct cache_set *c, const struct bkey *k)
 			size_t bucket = PTR_BUCKET_NR(c, k, i);
 			size_t r = bucket_remainder(c, PTR_OFFSET(k, i));
 
-			if (KEY_SIZE(k) + r > c->sb.bucket_size ||
+			if (KEY_SIZE(k) + r > c->cache->sb.bucket_size ||
 			    bucket <  ca->sb.first_bucket ||
 			    bucket >= ca->sb.nbuckets)
 				return true;
@@ -75,7 +75,7 @@ static const char *bch_ptr_status(struct cache_set *c, const struct bkey *k)
 			size_t bucket = PTR_BUCKET_NR(c, k, i);
 			size_t r = bucket_remainder(c, PTR_OFFSET(k, i));
 
-			if (KEY_SIZE(k) + r > c->sb.bucket_size)
+			if (KEY_SIZE(k) + r > c->cache->sb.bucket_size)
 				return "bad, length too big";
 			if (bucket <  ca->sb.first_bucket)
 				return "bad, short offset";
@@ -136,7 +136,7 @@ static void bch_bkey_dump(struct btree_keys *keys, const struct bkey *k)
 		size_t n = PTR_BUCKET_NR(b->c, k, j);
 
 		pr_cont(" bucket %zu", n);
-		if (n >= b->c->sb.first_bucket && n < b->c->sb.nbuckets)
+		if (n >= b->c->cache->sb.first_bucket && n < b->c->cache->sb.nbuckets)
 			pr_cont(" prio %i",
 				PTR_BUCKET(b->c, k, j)->prio);
 	}
diff --git a/drivers/md/bcache/features.c b/drivers/md/bcache/features.c
index 4442df48d28c..6469223f0b77 100644
--- a/drivers/md/bcache/features.c
+++ b/drivers/md/bcache/features.c
@@ -30,7 +30,7 @@ static struct feature feature_list[] = {
 	for (f = &feature_list[0]; f->compat != 0; f++) {		\
 		if (f->compat != BCH_FEATURE_ ## type)			\
 			continue;					\
-		if (BCH_HAS_ ## type ## _FEATURE(&c->sb, f->mask)) {	\
+		if (BCH_HAS_ ## type ## _FEATURE(&c->cache->sb, f->mask)) {	\
 			if (first) {					\
 				out += snprintf(out, buf + size - out,	\
 						"[");	\
@@ -44,7 +44,7 @@ static struct feature feature_list[] = {
 									\
 		out += snprintf(out, buf + size - out, "%s", f->string);\
 									\
-		if (BCH_HAS_ ## type ## _FEATURE(&c->sb, f->mask))	\
+		if (BCH_HAS_ ## type ## _FEATURE(&c->cache->sb, f->mask))	\
 			out += snprintf(out, buf + size - out, "]");	\
 									\
 		first = false;						\
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index a14a445618b4..dad71a6b7889 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -26,7 +26,7 @@ struct bio *bch_bbio_alloc(struct cache_set *c)
 	struct bbio *b = mempool_alloc(&c->bio_meta, GFP_NOIO);
 	struct bio *bio = &b->bio;
 
-	bio_init(bio, bio->bi_inline_vecs, meta_bucket_pages(&c->sb));
+	bio_init(bio, bio->bi_inline_vecs, meta_bucket_pages(&c->cache->sb));
 
 	return bio;
 }
diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c
index c1227bdb57e7..aefbdb7e003b 100644
--- a/drivers/md/bcache/journal.c
+++ b/drivers/md/bcache/journal.c
@@ -98,7 +98,7 @@ reread:		left = ca->sb.bucket_size - offset;
 				return ret;
 			}
 
-			blocks = set_blocks(j, block_bytes(ca->set));
+			blocks = set_blocks(j, block_bytes(ca));
 
 			/*
 			 * Nodes in 'list' are in linear increasing order of
@@ -179,112 +179,109 @@ int bch_journal_read(struct cache_set *c, struct list_head *list)
 		ret;							\
 	})
 
-	struct cache *ca;
-	unsigned int iter;
+	struct cache *ca = c->cache;
 	int ret = 0;
+	struct journal_device *ja = &ca->journal;
+	DECLARE_BITMAP(bitmap, SB_JOURNAL_BUCKETS);
+	unsigned int i, l, r, m;
+	uint64_t seq;
 
-	for_each_cache(ca, c, iter) {
-		struct journal_device *ja = &ca->journal;
-		DECLARE_BITMAP(bitmap, SB_JOURNAL_BUCKETS);
-		unsigned int i, l, r, m;
-		uint64_t seq;
-
-		bitmap_zero(bitmap, SB_JOURNAL_BUCKETS);
-		pr_debug("%u journal buckets\n", ca->sb.njournal_buckets);
+	bitmap_zero(bitmap, SB_JOURNAL_BUCKETS);
+	pr_debug("%u journal buckets\n", ca->sb.njournal_buckets);
 
+	/*
+	 * Read journal buckets ordered by golden ratio hash to quickly
+	 * find a sequence of buckets with valid journal entries
+	 */
+	for (i = 0; i < ca->sb.njournal_buckets; i++) {
 		/*
-		 * Read journal buckets ordered by golden ratio hash to quickly
-		 * find a sequence of buckets with valid journal entries
+		 * We must try the index l with ZERO first for
+		 * correctness due to the scenario that the journal
+		 * bucket is circular buffer which might have wrapped
 		 */
-		for (i = 0; i < ca->sb.njournal_buckets; i++) {
-			/*
-			 * We must try the index l with ZERO first for
-			 * correctness due to the scenario that the journal
-			 * bucket is circular buffer which might have wrapped
-			 */
-			l = (i * 2654435769U) % ca->sb.njournal_buckets;
+		l = (i * 2654435769U) % ca->sb.njournal_buckets;
 
-			if (test_bit(l, bitmap))
-				break;
+		if (test_bit(l, bitmap))
+			break;
 
-			if (read_bucket(l))
-				goto bsearch;
-		}
+		if (read_bucket(l))
+			goto bsearch;
+	}
 
-		/*
-		 * If that fails, check all the buckets we haven't checked
-		 * already
-		 */
-		pr_debug("falling back to linear search\n");
+	/*
+	 * If that fails, check all the buckets we haven't checked
+	 * already
+	 */
+	pr_debug("falling back to linear search\n");
 
-		for_each_clear_bit(l, bitmap, ca->sb.njournal_buckets)
-			if (read_bucket(l))
-				goto bsearch;
+	for_each_clear_bit(l, bitmap, ca->sb.njournal_buckets)
+		if (read_bucket(l))
+			goto bsearch;
 
-		/* no journal entries on this device? */
-		if (l == ca->sb.njournal_buckets)
-			continue;
+	/* no journal entries on this device? */
+	if (l == ca->sb.njournal_buckets)
+		goto out;
 bsearch:
-		BUG_ON(list_empty(list));
+	BUG_ON(list_empty(list));
 
-		/* Binary search */
-		m = l;
-		r = find_next_bit(bitmap, ca->sb.njournal_buckets, l + 1);
-		pr_debug("starting binary search, l %u r %u\n", l, r);
+	/* Binary search */
+	m = l;
+	r = find_next_bit(bitmap, ca->sb.njournal_buckets, l + 1);
+	pr_debug("starting binary search, l %u r %u\n", l, r);
 
-		while (l + 1 < r) {
-			seq = list_entry(list->prev, struct journal_replay,
-					 list)->j.seq;
+	while (l + 1 < r) {
+		seq = list_entry(list->prev, struct journal_replay,
+				 list)->j.seq;
 
-			m = (l + r) >> 1;
-			read_bucket(m);
+		m = (l + r) >> 1;
+		read_bucket(m);
 
-			if (seq != list_entry(list->prev, struct journal_replay,
-					      list)->j.seq)
-				l = m;
-			else
-				r = m;
-		}
+		if (seq != list_entry(list->prev, struct journal_replay,
+				      list)->j.seq)
+			l = m;
+		else
+			r = m;
+	}
 
-		/*
-		 * Read buckets in reverse order until we stop finding more
-		 * journal entries
-		 */
-		pr_debug("finishing up: m %u njournal_buckets %u\n",
-			 m, ca->sb.njournal_buckets);
-		l = m;
+	/*
+	 * Read buckets in reverse order until we stop finding more
+	 * journal entries
+	 */
+	pr_debug("finishing up: m %u njournal_buckets %u\n",
+		 m, ca->sb.njournal_buckets);
+	l = m;
 
-		while (1) {
-			if (!l--)
-				l = ca->sb.njournal_buckets - 1;
+	while (1) {
+		if (!l--)
+			l = ca->sb.njournal_buckets - 1;
 
-			if (l == m)
-				break;
+		if (l == m)
+			break;
 
-			if (test_bit(l, bitmap))
-				continue;
+		if (test_bit(l, bitmap))
+			continue;
 
-			if (!read_bucket(l))
-				break;
-		}
+		if (!read_bucket(l))
+			break;
+	}
 
-		seq = 0;
+	seq = 0;
 
-		for (i = 0; i < ca->sb.njournal_buckets; i++)
-			if (ja->seq[i] > seq) {
-				seq = ja->seq[i];
-				/*
-				 * When journal_reclaim() goes to allocate for
-				 * the first time, it'll use the bucket after
-				 * ja->cur_idx
-				 */
-				ja->cur_idx = i;
-				ja->last_idx = ja->discard_idx = (i + 1) %
-					ca->sb.njournal_buckets;
+	for (i = 0; i < ca->sb.njournal_buckets; i++)
+		if (ja->seq[i] > seq) {
+			seq = ja->seq[i];
+			/*
+			 * When journal_reclaim() goes to allocate for
+			 * the first time, it'll use the bucket after
+			 * ja->cur_idx
+			 */
+			ja->cur_idx = i;
+			ja->last_idx = ja->discard_idx = (i + 1) %
+				ca->sb.njournal_buckets;
 
-			}
-	}
+		}
 
+out:
 	if (!list_empty(list))
 		c->journal.seq = list_entry(list->prev,
 					    struct journal_replay,
@@ -342,12 +339,10 @@ void bch_journal_mark(struct cache_set *c, struct list_head *list)
 
 static bool is_discard_enabled(struct cache_set *s)
 {
-	struct cache *ca;
-	unsigned int i;
+	struct cache *ca = s->cache;
 
-	for_each_cache(ca, s, i)
-		if (ca->discard)
-			return true;
+	if (ca->discard)
+		return true;
 
 	return false;
 }
@@ -633,9 +628,10 @@ static void do_journal_discard(struct cache *ca)
 static void journal_reclaim(struct cache_set *c)
 {
 	struct bkey *k = &c->journal.key;
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	uint64_t last_seq;
-	unsigned int iter, n = 0;
+	unsigned int next;
+	struct journal_device *ja = &ca->journal;
 	atomic_t p __maybe_unused;
 
 	atomic_long_inc(&c->reclaim);
@@ -647,46 +643,31 @@ static void journal_reclaim(struct cache_set *c)
 
 	/* Update last_idx */
 
-	for_each_cache(ca, c, iter) {
-		struct journal_device *ja = &ca->journal;
-
-		while (ja->last_idx != ja->cur_idx &&
-		       ja->seq[ja->last_idx] < last_seq)
-			ja->last_idx = (ja->last_idx + 1) %
-				ca->sb.njournal_buckets;
-	}
+	while (ja->last_idx != ja->cur_idx &&
+	       ja->seq[ja->last_idx] < last_seq)
+		ja->last_idx = (ja->last_idx + 1) %
+			ca->sb.njournal_buckets;
 
-	for_each_cache(ca, c, iter)
-		do_journal_discard(ca);
+	do_journal_discard(ca);
 
 	if (c->journal.blocks_free)
 		goto out;
 
-	/*
-	 * Allocate:
-	 * XXX: Sort by free journal space
-	 */
-
-	for_each_cache(ca, c, iter) {
-		struct journal_device *ja = &ca->journal;
-		unsigned int next = (ja->cur_idx + 1) % ca->sb.njournal_buckets;
+	next = (ja->cur_idx + 1) % ca->sb.njournal_buckets;
+	/* No space available on this device */
+	if (next == ja->discard_idx)
+		goto out;
 
-		/* No space available on this device */
-		if (next == ja->discard_idx)
-			continue;
+	ja->cur_idx = next;
+	k->ptr[0] = MAKE_PTR(0,
+			     bucket_to_sector(c, ca->sb.d[ja->cur_idx]),
+			     ca->sb.nr_this_dev);
+	atomic_long_inc(&c->reclaimed_journal_buckets);
 
-		ja->cur_idx = next;
-		k->ptr[n++] = MAKE_PTR(0,
-				  bucket_to_sector(c, ca->sb.d[ja->cur_idx]),
-				  ca->sb.nr_this_dev);
-		atomic_long_inc(&c->reclaimed_journal_buckets);
-	}
+	bkey_init(k);
+	SET_KEY_PTRS(k, 1);
+	c->journal.blocks_free = ca->sb.bucket_size >> c->block_bits;
 
-	if (n) {
-		bkey_init(k);
-		SET_KEY_PTRS(k, n);
-		c->journal.blocks_free = c->sb.bucket_size >> c->block_bits;
-	}
 out:
 	if (!journal_full(&c->journal))
 		__closure_wake_up(&c->journal.wait);
@@ -750,11 +731,11 @@ static void journal_write_unlocked(struct closure *cl)
 	__releases(c->journal.lock)
 {
 	struct cache_set *c = container_of(cl, struct cache_set, journal.io);
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	struct journal_write *w = c->journal.cur;
 	struct bkey *k = &c->journal.key;
-	unsigned int i, sectors = set_blocks(w->data, block_bytes(c)) *
-		c->sb.block_size;
+	unsigned int i, sectors = set_blocks(w->data, block_bytes(ca)) *
+		ca->sb.block_size;
 
 	struct bio *bio;
 	struct bio_list list;
@@ -773,17 +754,15 @@ static void journal_write_unlocked(struct closure *cl)
 		return;
 	}
 
-	c->journal.blocks_free -= set_blocks(w->data, block_bytes(c));
+	c->journal.blocks_free -= set_blocks(w->data, block_bytes(ca));
 
 	w->data->btree_level = c->root->level;
 
 	bkey_copy(&w->data->btree_root, &c->root->key);
 	bkey_copy(&w->data->uuid_bucket, &c->uuid_bucket);
 
-	for_each_cache(ca, c, i)
-		w->data->prio_bucket[ca->sb.nr_this_dev] = ca->prio_buckets[0];
-
-	w->data->magic		= jset_magic(&c->sb);
+	w->data->prio_bucket[ca->sb.nr_this_dev] = ca->prio_buckets[0];
+	w->data->magic		= jset_magic(&ca->sb);
 	w->data->version	= BCACHE_JSET_VERSION;
 	w->data->last_seq	= last_seq(&c->journal);
 	w->data->csum		= csum_set(w->data);
@@ -859,6 +838,7 @@ static struct journal_write *journal_wait_for_write(struct cache_set *c,
 	size_t sectors;
 	struct closure cl;
 	bool wait = false;
+	struct cache *ca = c->cache;
 
 	closure_init_stack(&cl);
 
@@ -868,10 +848,10 @@ static struct journal_write *journal_wait_for_write(struct cache_set *c,
 		struct journal_write *w = c->journal.cur;
 
 		sectors = __set_blocks(w->data, w->data->keys + nkeys,
-				       block_bytes(c)) * c->sb.block_size;
+				       block_bytes(ca)) * ca->sb.block_size;
 
 		if (sectors <= min_t(size_t,
-				     c->journal.blocks_free * c->sb.block_size,
+				     c->journal.blocks_free * ca->sb.block_size,
 				     PAGE_SECTORS << JSET_BITS))
 			return w;
 
@@ -936,7 +916,7 @@ atomic_t *bch_journal(struct cache_set *c,
 	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags)))
 		return NULL;
 
-	if (!CACHE_SYNC(&c->sb))
+	if (!CACHE_SYNC(&c->cache->sb))
 		return NULL;
 
 	w = journal_wait_for_write(c, bch_keylist_nkeys(keys));
diff --git a/drivers/md/bcache/movinggc.c b/drivers/md/bcache/movinggc.c
index 5872d6470470..b9c3d27ec093 100644
--- a/drivers/md/bcache/movinggc.c
+++ b/drivers/md/bcache/movinggc.c
@@ -196,50 +196,48 @@ static unsigned int bucket_heap_top(struct cache *ca)
 
 void bch_moving_gc(struct cache_set *c)
 {
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	struct bucket *b;
-	unsigned int i;
+	unsigned long sectors_to_move, reserve_sectors;
 
 	if (!c->copy_gc_enabled)
 		return;
 
 	mutex_lock(&c->bucket_lock);
 
-	for_each_cache(ca, c, i) {
-		unsigned long sectors_to_move = 0;
-		unsigned long reserve_sectors = ca->sb.bucket_size *
+	sectors_to_move = 0;
+	reserve_sectors = ca->sb.bucket_size *
 			     fifo_used(&ca->free[RESERVE_MOVINGGC]);
 
-		ca->heap.used = 0;
-
-		for_each_bucket(b, ca) {
-			if (GC_MARK(b) == GC_MARK_METADATA ||
-			    !GC_SECTORS_USED(b) ||
-			    GC_SECTORS_USED(b) == ca->sb.bucket_size ||
-			    atomic_read(&b->pin))
-				continue;
-
-			if (!heap_full(&ca->heap)) {
-				sectors_to_move += GC_SECTORS_USED(b);
-				heap_add(&ca->heap, b, bucket_cmp);
-			} else if (bucket_cmp(b, heap_peek(&ca->heap))) {
-				sectors_to_move -= bucket_heap_top(ca);
-				sectors_to_move += GC_SECTORS_USED(b);
-
-				ca->heap.data[0] = b;
-				heap_sift(&ca->heap, 0, bucket_cmp);
-			}
-		}
+	ca->heap.used = 0;
+
+	for_each_bucket(b, ca) {
+		if (GC_MARK(b) == GC_MARK_METADATA ||
+		    !GC_SECTORS_USED(b) ||
+		    GC_SECTORS_USED(b) == ca->sb.bucket_size ||
+		    atomic_read(&b->pin))
+			continue;
 
-		while (sectors_to_move > reserve_sectors) {
-			heap_pop(&ca->heap, b, bucket_cmp);
-			sectors_to_move -= GC_SECTORS_USED(b);
+		if (!heap_full(&ca->heap)) {
+			sectors_to_move += GC_SECTORS_USED(b);
+			heap_add(&ca->heap, b, bucket_cmp);
+		} else if (bucket_cmp(b, heap_peek(&ca->heap))) {
+			sectors_to_move -= bucket_heap_top(ca);
+			sectors_to_move += GC_SECTORS_USED(b);
+
+			ca->heap.data[0] = b;
+			heap_sift(&ca->heap, 0, bucket_cmp);
 		}
+	}
 
-		while (heap_pop(&ca->heap, b, bucket_cmp))
-			SET_GC_MOVE(b, 1);
+	while (sectors_to_move > reserve_sectors) {
+		heap_pop(&ca->heap, b, bucket_cmp);
+		sectors_to_move -= GC_SECTORS_USED(b);
 	}
 
+	while (heap_pop(&ca->heap, b, bucket_cmp))
+		SET_GC_MOVE(b, 1);
+
 	mutex_unlock(&c->bucket_lock);
 
 	c->moving_gc_keys.last_scanned = ZERO_KEY;
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index 7f54ae223644..214326383145 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -99,7 +99,7 @@ static int bch_keylist_realloc(struct keylist *l, unsigned int u64s,
 	 * bch_data_insert_keys() will insert the keys created so far
 	 * and finish the rest when the keylist is empty.
 	 */
-	if (newsize * sizeof(uint64_t) > block_bytes(c) - sizeof(struct jset))
+	if (newsize * sizeof(uint64_t) > block_bytes(c->cache) - sizeof(struct jset))
 		return -ENOMEM;
 
 	return __bch_keylist_realloc(l, u64s);
@@ -394,8 +394,8 @@ static bool check_should_bypass(struct cached_dev *dc, struct bio *bio)
 			goto skip;
 	}
 
-	if (bio->bi_iter.bi_sector & (c->sb.block_size - 1) ||
-	    bio_sectors(bio) & (c->sb.block_size - 1)) {
+	if (bio->bi_iter.bi_sector & (c->cache->sb.block_size - 1) ||
+	    bio_sectors(bio) & (c->cache->sb.block_size - 1)) {
 		pr_debug("skipping unaligned io\n");
 		goto skip;
 	}
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 6bfa77167362..46a00134a36a 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -343,34 +343,25 @@ static void bcache_write_super_unlock(struct closure *cl)
 void bcache_write_super(struct cache_set *c)
 {
 	struct closure *cl = &c->sb_write;
-	struct cache *ca;
-	unsigned int i, version = BCACHE_SB_VERSION_CDEV_WITH_UUID;
+	struct cache *ca = c->cache;
+	struct bio *bio = &ca->sb_bio;
+	unsigned int version = BCACHE_SB_VERSION_CDEV_WITH_UUID;
 
 	down(&c->sb_write_mutex);
 	closure_init(cl, &c->cl);
 
-	c->sb.seq++;
-
-	if (c->sb.version > version)
-		version = c->sb.version;
-
-	for_each_cache(ca, c, i) {
-		struct bio *bio = &ca->sb_bio;
-
-		ca->sb.version		= version;
-		ca->sb.seq		= c->sb.seq;
-		ca->sb.last_mount	= c->sb.last_mount;
+	ca->sb.seq++;
 
-		SET_CACHE_SYNC(&ca->sb, CACHE_SYNC(&c->sb));
+	if (ca->sb.version < version)
+		ca->sb.version = version;
 
-		bio_init(bio, ca->sb_bv, 1);
-		bio_set_dev(bio, ca->bdev);
-		bio->bi_end_io	= write_super_endio;
-		bio->bi_private = ca;
+	bio_init(bio, ca->sb_bv, 1);
+	bio_set_dev(bio, ca->bdev);
+	bio->bi_end_io	= write_super_endio;
+	bio->bi_private = ca;
 
-		closure_get(cl);
-		__write_super(&ca->sb, ca->sb_disk, bio);
-	}
+	closure_get(cl);
+	__write_super(&ca->sb, ca->sb_disk, bio);
 
 	closure_return_with_destructor(cl, bcache_write_super_unlock);
 }
@@ -480,22 +471,21 @@ static int __uuid_write(struct cache_set *c)
 {
 	BKEY_PADDED(key) k;
 	struct closure cl;
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	unsigned int size;
 
 	closure_init_stack(&cl);
 	lockdep_assert_held(&bch_register_lock);
 
-	if (bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, 1, true))
+	if (bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, true))
 		return 1;
 
-	size =  meta_bucket_pages(&c->sb) * PAGE_SECTORS;
+	size =  meta_bucket_pages(&ca->sb) * PAGE_SECTORS;
 	SET_KEY_SIZE(&k.key, size);
 	uuid_io(c, REQ_OP_WRITE, 0, &k.key, &cl);
 	closure_sync(&cl);
 
 	/* Only one bucket used for uuid write */
-	ca = PTR_CACHE(c, &k.key, 0);
 	atomic_long_add(ca->sb.bucket_size, &ca->meta_sectors_written);
 
 	bkey_copy(&c->uuid_bucket, &k.key);
@@ -772,26 +762,22 @@ static void bcache_device_unlink(struct bcache_device *d)
 	lockdep_assert_held(&bch_register_lock);
 
 	if (d->c && !test_and_set_bit(BCACHE_DEV_UNLINK_DONE, &d->flags)) {
-		unsigned int i;
-		struct cache *ca;
+		struct cache *ca = d->c->cache;
 
 		sysfs_remove_link(&d->c->kobj, d->name);
 		sysfs_remove_link(&d->kobj, "cache");
 
-		for_each_cache(ca, d->c, i)
-			bd_unlink_disk_holder(ca->bdev, d->disk);
+		bd_unlink_disk_holder(ca->bdev, d->disk);
 	}
 }
 
 static void bcache_device_link(struct bcache_device *d, struct cache_set *c,
 			       const char *name)
 {
-	unsigned int i;
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	int ret;
 
-	for_each_cache(ca, d->c, i)
-		bd_link_disk_holder(ca->bdev, d->disk);
+	bd_link_disk_holder(ca->bdev, d->disk);
 
 	snprintf(d->name, BCACHEDEVNAME_SIZE,
 		 "%s%u", name, d->id);
@@ -1196,8 +1182,8 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
 	struct cached_dev *exist_dc, *t;
 	int ret = 0;
 
-	if ((set_uuid && memcmp(set_uuid, c->sb.set_uuid, 16)) ||
-	    (!set_uuid && memcmp(dc->sb.set_uuid, c->sb.set_uuid, 16)))
+	if ((set_uuid && memcmp(set_uuid, c->set_uuid, 16)) ||
+	    (!set_uuid && memcmp(dc->sb.set_uuid, c->set_uuid, 16)))
 		return -ENOENT;
 
 	if (dc->disk.c) {
@@ -1212,7 +1198,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
 		return -EINVAL;
 	}
 
-	if (dc->sb.block_size < c->sb.block_size) {
+	if (dc->sb.block_size < c->cache->sb.block_size) {
 		/* Will die */
 		pr_err("Couldn't attach %s: block size less than set's block size\n",
 		       dc->backing_dev_name);
@@ -1269,7 +1255,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
 		u->first_reg = u->last_reg = rtime;
 		bch_uuid_write(c);
 
-		memcpy(dc->sb.set_uuid, c->sb.set_uuid, 16);
+		memcpy(dc->sb.set_uuid, c->set_uuid, 16);
 		SET_BDEV_STATE(&dc->sb, BDEV_STATE_CLEAN);
 
 		bch_write_bdev_super(dc, &cl);
@@ -1331,7 +1317,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
 	pr_info("Caching %s as %s on set %pU\n",
 		dc->backing_dev_name,
 		dc->disk.disk->disk_name,
-		dc->disk.c->sb.set_uuid);
+		dc->disk.c->set_uuid);
 	return 0;
 }
 
@@ -1534,7 +1520,7 @@ static int flash_dev_run(struct cache_set *c, struct uuid_entry *u)
 
 	kobject_init(&d->kobj, &bch_flash_dev_ktype);
 
-	if (bcache_device_init(d, block_bytes(c), u->sectors,
+	if (bcache_device_init(d, block_bytes(c->cache), u->sectors,
 			NULL, &bcache_flash_ops))
 		goto err;
 
@@ -1638,7 +1624,7 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...)
 	vaf.va = &args;
 
 	pr_err("error on %pU: %pV, disabling caching\n",
-	       c->sb.set_uuid, &vaf);
+	       c->set_uuid, &vaf);
 
 	va_end(args);
 
@@ -1662,7 +1648,6 @@ static void cache_set_free(struct closure *cl)
 {
 	struct cache_set *c = container_of(cl, struct cache_set, cl);
 	struct cache *ca;
-	unsigned int i;
 
 	debugfs_remove(c->debug);
 
@@ -1671,15 +1656,16 @@ static void cache_set_free(struct closure *cl)
 	bch_journal_free(c);
 
 	mutex_lock(&bch_register_lock);
-	for_each_cache(ca, c, i)
-		if (ca) {
-			ca->set = NULL;
-			c->cache[ca->sb.nr_this_dev] = NULL;
-			kobject_put(&ca->kobj);
-		}
-
 	bch_bset_sort_state_free(&c->sort);
-	free_pages((unsigned long) c->uuids, ilog2(meta_bucket_pages(&c->sb)));
+	free_pages((unsigned long) c->uuids, ilog2(meta_bucket_pages(&c->cache->sb)));
+
+	ca = c->cache;
+	if (ca) {
+		ca->set = NULL;
+		c->cache = NULL;
+		kobject_put(&ca->kobj);
+	}
+
 
 	if (c->moving_gc_wq)
 		destroy_workqueue(c->moving_gc_wq);
@@ -1692,7 +1678,7 @@ static void cache_set_free(struct closure *cl)
 	list_del(&c->list);
 	mutex_unlock(&bch_register_lock);
 
-	pr_info("Cache set %pU unregistered\n", c->sb.set_uuid);
+	pr_info("Cache set %pU unregistered\n", c->set_uuid);
 	wake_up(&unregister_wait);
 
 	closure_debug_destroy(&c->cl);
@@ -1702,9 +1688,8 @@ static void cache_set_free(struct closure *cl)
 static void cache_set_flush(struct closure *cl)
 {
 	struct cache_set *c = container_of(cl, struct cache_set, caching);
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	struct btree *b;
-	unsigned int i;
 
 	bch_cache_accounting_destroy(&c->accounting);
 
@@ -1729,9 +1714,8 @@ static void cache_set_flush(struct closure *cl)
 			mutex_unlock(&b->write_lock);
 		}
 
-	for_each_cache(ca, c, i)
-		if (ca->alloc_thread)
-			kthread_stop(ca->alloc_thread);
+	if (ca->alloc_thread)
+		kthread_stop(ca->alloc_thread);
 
 	if (c->journal.cur) {
 		cancel_delayed_work_sync(&c->journal.work);
@@ -1764,7 +1748,7 @@ static void conditional_stop_bcache_device(struct cache_set *c,
 {
 	if (dc->stop_when_cache_set_failed == BCH_CACHED_DEV_STOP_ALWAYS) {
 		pr_warn("stop_when_cache_set_failed of %s is \"always\", stop it for failed cache set %pU.\n",
-			d->disk->disk_name, c->sb.set_uuid);
+			d->disk->disk_name, c->set_uuid);
 		bcache_device_stop(d);
 	} else if (atomic_read(&dc->has_dirty)) {
 		/*
@@ -1841,15 +1825,13 @@ void bch_cache_set_unregister(struct cache_set *c)
 	bch_cache_set_stop(c);
 }
 
-#define alloc_bucket_pages(gfp, c)			\
-	((void *) __get_free_pages(__GFP_ZERO|__GFP_COMP|gfp, ilog2(bucket_pages(c))))
-
 #define alloc_meta_bucket_pages(gfp, sb)		\
 	((void *) __get_free_pages(__GFP_ZERO|__GFP_COMP|gfp, ilog2(meta_bucket_pages(sb))))
 
 struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 {
 	int iter_size;
+	struct cache *ca = container_of(sb, struct cache, sb);
 	struct cache_set *c = kzalloc(sizeof(struct cache_set), GFP_KERNEL);
 
 	if (!c)
@@ -1871,24 +1853,16 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 
 	bch_cache_accounting_init(&c->accounting, &c->cl);
 
-	memcpy(c->sb.set_uuid, sb->set_uuid, 16);
-	c->sb.block_size	= sb->block_size;
-	c->sb.bucket_size	= sb->bucket_size;
-	c->sb.nr_in_set		= sb->nr_in_set;
-	c->sb.last_mount	= sb->last_mount;
-	c->sb.version		= sb->version;
-	if (c->sb.version >= BCACHE_SB_VERSION_CDEV_WITH_FEATURES) {
-		c->sb.feature_compat = sb->feature_compat;
-		c->sb.feature_ro_compat = sb->feature_ro_compat;
-		c->sb.feature_incompat = sb->feature_incompat;
-	}
+	memcpy(c->set_uuid, sb->set_uuid, 16);
 
+	c->cache		= ca;
+	c->cache->set		= c;
 	c->bucket_bits		= ilog2(sb->bucket_size);
 	c->block_bits		= ilog2(sb->block_size);
-	c->nr_uuids		= meta_bucket_bytes(&c->sb) / sizeof(struct uuid_entry);
+	c->nr_uuids		= meta_bucket_bytes(sb) / sizeof(struct uuid_entry);
 	c->devices_max_used	= 0;
 	atomic_set(&c->attached_dev_nr, 0);
-	c->btree_pages		= meta_bucket_pages(&c->sb);
+	c->btree_pages		= meta_bucket_pages(sb);
 	if (c->btree_pages > BTREE_MAX_PAGES)
 		c->btree_pages = max_t(int, c->btree_pages / 4,
 				       BTREE_MAX_PAGES);
@@ -1926,7 +1900,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 
 	if (mempool_init_kmalloc_pool(&c->bio_meta, 2,
 			sizeof(struct bbio) +
-			sizeof(struct bio_vec) * meta_bucket_pages(&c->sb)))
+			sizeof(struct bio_vec) * meta_bucket_pages(sb)))
 		goto err;
 
 	if (mempool_init_kmalloc_pool(&c->fill_iter, 1, iter_size))
@@ -1936,7 +1910,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 			BIOSET_NEED_BVECS|BIOSET_NEED_RESCUER))
 		goto err;
 
-	c->uuids = alloc_meta_bucket_pages(GFP_KERNEL, &c->sb);
+	c->uuids = alloc_meta_bucket_pages(GFP_KERNEL, sb);
 	if (!c->uuids)
 		goto err;
 
@@ -1972,19 +1946,17 @@ static int run_cache_set(struct cache_set *c)
 {
 	const char *err = "cannot allocate memory";
 	struct cached_dev *dc, *t;
-	struct cache *ca;
+	struct cache *ca = c->cache;
 	struct closure cl;
-	unsigned int i;
 	LIST_HEAD(journal);
 	struct journal_replay *l;
 
 	closure_init_stack(&cl);
 
-	for_each_cache(ca, c, i)
-		c->nbuckets += ca->sb.nbuckets;
+	c->nbuckets = ca->sb.nbuckets;
 	set_gc_sectors(c);
 
-	if (CACHE_SYNC(&c->sb)) {
+	if (CACHE_SYNC(&c->cache->sb)) {
 		struct bkey *k;
 		struct jset *j;
 
@@ -2001,10 +1973,8 @@ static int run_cache_set(struct cache_set *c)
 		j = &list_entry(journal.prev, struct journal_replay, list)->j;
 
 		err = "IO error reading priorities";
-		for_each_cache(ca, c, i) {
-			if (prio_read(ca, j->prio_bucket[ca->sb.nr_this_dev]))
-				goto err;
-		}
+		if (prio_read(ca, j->prio_bucket[ca->sb.nr_this_dev]))
+			goto err;
 
 		/*
 		 * If prio_read() fails it'll call cache_set_error and we'll
@@ -2048,9 +2018,8 @@ static int run_cache_set(struct cache_set *c)
 		bch_journal_next(&c->journal);
 
 		err = "error starting allocator thread";
-		for_each_cache(ca, c, i)
-			if (bch_cache_allocator_start(ca))
-				goto err;
+		if (bch_cache_allocator_start(ca))
+			goto err;
 
 		/*
 		 * First place it's safe to allocate: btree_check() and
@@ -2069,28 +2038,23 @@ static int run_cache_set(struct cache_set *c)
 		if (bch_journal_replay(c, &journal))
 			goto err;
 	} else {
-		pr_notice("invalidating existing data\n");
+		unsigned int j;
 
-		for_each_cache(ca, c, i) {
-			unsigned int j;
+		pr_notice("invalidating existing data\n");
+		ca->sb.keys = clamp_t(int, ca->sb.nbuckets >> 7,
+					2, SB_JOURNAL_BUCKETS);
 
-			ca->sb.keys = clamp_t(int, ca->sb.nbuckets >> 7,
-					      2, SB_JOURNAL_BUCKETS);
-
-			for (j = 0; j < ca->sb.keys; j++)
-				ca->sb.d[j] = ca->sb.first_bucket + j;
-		}
+		for (j = 0; j < ca->sb.keys; j++)
+			ca->sb.d[j] = ca->sb.first_bucket + j;
 
 		bch_initial_gc_finish(c);
 
 		err = "error starting allocator thread";
-		for_each_cache(ca, c, i)
-			if (bch_cache_allocator_start(ca))
-				goto err;
+		if (bch_cache_allocator_start(ca))
+			goto err;
 
 		mutex_lock(&c->bucket_lock);
-		for_each_cache(ca, c, i)
-			bch_prio_write(ca, true);
+		bch_prio_write(ca, true);
 		mutex_unlock(&c->bucket_lock);
 
 		err = "cannot allocate new UUID bucket";
@@ -2115,7 +2079,7 @@ static int run_cache_set(struct cache_set *c)
 		 * everything is set up - fortunately journal entries won't be
 		 * written until the SET_CACHE_SYNC() here:
 		 */
-		SET_CACHE_SYNC(&c->sb, true);
+		SET_CACHE_SYNC(&c->cache->sb, true);
 
 		bch_journal_next(&c->journal);
 		bch_journal_meta(c, &cl);
@@ -2126,7 +2090,7 @@ static int run_cache_set(struct cache_set *c)
 		goto err;
 
 	closure_sync(&cl);
-	c->sb.last_mount = (u32)ktime_get_real_seconds();
+	c->cache->sb.last_mount = (u32)ktime_get_real_seconds();
 	bcache_write_super(c);
 
 	list_for_each_entry_safe(dc, t, &uncached_devices, list)
@@ -2150,13 +2114,6 @@ err:
 	return -EIO;
 }
 
-static bool can_attach_cache(struct cache *ca, struct cache_set *c)
-{
-	return ca->sb.block_size	== c->sb.block_size &&
-		ca->sb.bucket_size	== c->sb.bucket_size &&
-		ca->sb.nr_in_set	== c->sb.nr_in_set;
-}
-
 static const char *register_cache_set(struct cache *ca)
 {
 	char buf[12];
@@ -2164,16 +2121,10 @@ static const char *register_cache_set(struct cache *ca)
 	struct cache_set *c;
 
 	list_for_each_entry(c, &bch_cache_sets, list)
-		if (!memcmp(c->sb.set_uuid, ca->sb.set_uuid, 16)) {
-			if (c->cache[ca->sb.nr_this_dev])
+		if (!memcmp(c->set_uuid, ca->sb.set_uuid, 16)) {
+			if (c->cache)
 				return "duplicate cache set member";
 
-			if (!can_attach_cache(ca, c))
-				return "cache sb does not match set";
-
-			if (!CACHE_SYNC(&ca->sb))
-				SET_CACHE_SYNC(&c->sb, false);
-
 			goto found;
 		}
 
@@ -2182,7 +2133,7 @@ static const char *register_cache_set(struct cache *ca)
 		return err;
 
 	err = "error creating kobject";
-	if (kobject_add(&c->kobj, bcache_kobj, "%pU", c->sb.set_uuid) ||
+	if (kobject_add(&c->kobj, bcache_kobj, "%pU", c->set_uuid) ||
 	    kobject_add(&c->internal, &c->kobj, "internal"))
 		goto err;
 
@@ -2198,31 +2149,13 @@ found:
 	    sysfs_create_link(&c->kobj, &ca->kobj, buf))
 		goto err;
 
-	/*
-	 * A special case is both ca->sb.seq and c->sb.seq are 0,
-	 * such condition happens on a new created cache device whose
-	 * super block is never flushed yet. In this case c->sb.version
-	 * and other members should be updated too, otherwise we will
-	 * have a mistaken super block version in cache set.
-	 */
-	if (ca->sb.seq > c->sb.seq || c->sb.seq == 0) {
-		c->sb.version		= ca->sb.version;
-		memcpy(c->sb.set_uuid, ca->sb.set_uuid, 16);
-		c->sb.flags             = ca->sb.flags;
-		c->sb.seq		= ca->sb.seq;
-		pr_debug("set version = %llu\n", c->sb.version);
-	}
-
 	kobject_get(&ca->kobj);
 	ca->set = c;
-	ca->set->cache[ca->sb.nr_this_dev] = ca;
-	c->cache_by_alloc[c->caches_loaded++] = ca;
+	ca->set->cache = ca;
 
-	if (c->caches_loaded == c->sb.nr_in_set) {
-		err = "failed to run cache set";
-		if (run_cache_set(c) < 0)
-			goto err;
-	}
+	err = "failed to run cache set";
+	if (run_cache_set(c) < 0)
+		goto err;
 
 	return NULL;
 err:
@@ -2239,8 +2172,8 @@ void bch_cache_release(struct kobject *kobj)
 	unsigned int i;
 
 	if (ca->set) {
-		BUG_ON(ca->set->cache[ca->sb.nr_this_dev] != ca);
-		ca->set->cache[ca->sb.nr_this_dev] = NULL;
+		BUG_ON(ca->set->cache != ca);
+		ca->set->cache = NULL;
 	}
 
 	free_pages((unsigned long) ca->disk_buckets, ilog2(meta_bucket_pages(&ca->sb)));
@@ -2448,7 +2381,6 @@ static ssize_t bch_pending_bdevs_cleanup(struct kobject *k,
 
 kobj_attribute_write(register,		register_bcache);
 kobj_attribute_write(register_quiet,	register_bcache);
-kobj_attribute_write(register_async,	register_bcache);
 kobj_attribute_write(pendings_cleanup,	bch_pending_bdevs_cleanup);
 
 static bool bch_is_open_backing(struct block_device *bdev)
@@ -2469,13 +2401,14 @@ static bool bch_is_open_backing(struct block_device *bdev)
 static bool bch_is_open_cache(struct block_device *bdev)
 {
 	struct cache_set *c, *tc;
-	struct cache *ca;
-	unsigned int i;
 
-	list_for_each_entry_safe(c, tc, &bch_cache_sets, list)
-		for_each_cache(ca, c, i)
-			if (ca->bdev == bdev)
-				return true;
+	list_for_each_entry_safe(c, tc, &bch_cache_sets, list) {
+		struct cache *ca = c->cache;
+
+		if (ca->bdev == bdev)
+			return true;
+	}
+
 	return false;
 }
 
@@ -2571,6 +2504,11 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
 	struct cache_sb_disk *sb_disk;
 	struct block_device *bdev;
 	ssize_t ret;
+	bool async_registration = false;
+
+#ifdef CONFIG_BCACHE_ASYNC_REGISTRATION
+	async_registration = true;
+#endif
 
 	ret = -EBUSY;
 	err = "failed to reference bcache module";
@@ -2624,7 +2562,8 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
 		goto out_blkdev_put;
 
 	err = "failed to register device";
-	if (attr == &ksysfs_register_async) {
+
+	if (async_registration) {
 		/* register in asynchronous way */
 		struct async_reg_args *args =
 			kzalloc(sizeof(struct async_reg_args), GFP_KERNEL);
@@ -2719,7 +2658,7 @@ static ssize_t bch_pending_bdevs_cleanup(struct kobject *k,
 	list_for_each_entry_safe(pdev, tpdev, &pending_devs, list) {
 		list_for_each_entry_safe(c, tc, &bch_cache_sets, list) {
 			char *pdev_set_uuid = pdev->dc->sb.set_uuid;
-			char *set_uuid = c->sb.uuid;
+			char *set_uuid = c->set_uuid;
 
 			if (!memcmp(pdev_set_uuid, set_uuid, 16)) {
 				list_del(&pdev->list);
@@ -2887,9 +2826,6 @@ static int __init bcache_init(void)
 	static const struct attribute *files[] = {
 		&ksysfs_register.attr,
 		&ksysfs_register_quiet.attr,
-#ifdef CONFIG_BCACHE_ASYNC_REGISTRATION
-		&ksysfs_register_async.attr,
-#endif
 		&ksysfs_pendings_cleanup.attr,
 		NULL
 	};
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index ac06c0bc3c0a..554e3afc9b68 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -711,10 +711,10 @@ SHOW(__bch_cache_set)
 {
 	struct cache_set *c = container_of(kobj, struct cache_set, kobj);
 
-	sysfs_print(synchronous,		CACHE_SYNC(&c->sb));
+	sysfs_print(synchronous,		CACHE_SYNC(&c->cache->sb));
 	sysfs_print(journal_delay_ms,		c->journal_delay_ms);
-	sysfs_hprint(bucket_size,		bucket_bytes(c));
-	sysfs_hprint(block_size,		block_bytes(c));
+	sysfs_hprint(bucket_size,		bucket_bytes(c->cache));
+	sysfs_hprint(block_size,		block_bytes(c->cache));
 	sysfs_print(tree_depth,			c->root->level);
 	sysfs_print(root_usage_percent,		bch_root_usage(c));
 
@@ -812,8 +812,8 @@ STORE(__bch_cache_set)
 	if (attr == &sysfs_synchronous) {
 		bool sync = strtoul_or_return(buf);
 
-		if (sync != CACHE_SYNC(&c->sb)) {
-			SET_CACHE_SYNC(&c->sb, sync);
+		if (sync != CACHE_SYNC(&c->cache->sb)) {
+			SET_CACHE_SYNC(&c->cache->sb, sync);
 			bcache_write_super(c);
 		}
 	}
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index 4f4ad6b3d43a..3c74996978da 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -35,7 +35,7 @@ static uint64_t __calc_target_rate(struct cached_dev *dc)
 	 * This is the size of the cache, minus the amount used for
 	 * flash-only devices
 	 */
-	uint64_t cache_sectors = c->nbuckets * c->sb.bucket_size -
+	uint64_t cache_sectors = c->nbuckets * c->cache->sb.bucket_size -
 				atomic_long_read(&c->flash_dev_dirty_sectors);
 
 	/*
diff --git a/drivers/md/md-bitmap.c b/drivers/md/md-bitmap.c
index b10c51988c8e..200c5d0f08bf 100644
--- a/drivers/md/md-bitmap.c
+++ b/drivers/md/md-bitmap.c
@@ -357,11 +357,12 @@ static int read_page(struct file *file, unsigned long index,
 	struct inode *inode = file_inode(file);
 	struct buffer_head *bh;
 	sector_t block, blk_cur;
+	unsigned long blocksize = i_blocksize(inode);
 
 	pr_debug("read bitmap file (%dB @ %llu)\n", (int)PAGE_SIZE,
 		 (unsigned long long)index << PAGE_SHIFT);
 
-	bh = alloc_page_buffers(page, 1<<inode->i_blkbits, false);
+	bh = alloc_page_buffers(page, blocksize, false);
 	if (!bh) {
 		ret = -ENOMEM;
 		goto out;
@@ -383,10 +384,10 @@ static int read_page(struct file *file, unsigned long index,
 
 			bh->b_blocknr = block;
 			bh->b_bdev = inode->i_sb->s_bdev;
-			if (count < (1<<inode->i_blkbits))
+			if (count < blocksize)
 				count = 0;
 			else
-				count -= (1<<inode->i_blkbits);
+				count -= blocksize;
 
 			bh->b_end_io = end_bitmap_write;
 			bh->b_private = bitmap;
@@ -605,8 +606,8 @@ re_read:
 	if (bitmap->cluster_slot >= 0) {
 		sector_t bm_blocks = bitmap->mddev->resync_max_sectors;
 
-		sector_div(bm_blocks,
-			   bitmap->mddev->bitmap_info.chunksize >> 9);
+		bm_blocks = DIV_ROUND_UP_SECTOR_T(bm_blocks,
+			   (bitmap->mddev->bitmap_info.chunksize >> 9));
 		/* bits to bytes */
 		bm_blocks = ((bm_blocks+7) >> 3) + sizeof(bitmap_super_t);
 		/* to 4k blocks */
@@ -1367,7 +1368,7 @@ __acquires(bitmap->lock)
 	if (bitmap->bp[page].hijacked ||
 	    bitmap->bp[page].map == NULL)
 		csize = ((sector_t)1) << (bitmap->chunkshift +
-					  PAGE_COUNTER_SHIFT - 1);
+					  PAGE_COUNTER_SHIFT);
 	else
 		csize = ((sector_t)1) << bitmap->chunkshift;
 	*blocks = csize - (offset & (csize - 1));
@@ -1949,6 +1950,7 @@ out:
 }
 EXPORT_SYMBOL_GPL(md_bitmap_load);
 
+/* caller need to free returned bitmap with md_bitmap_free() */
 struct bitmap *get_bitmap_from_slot(struct mddev *mddev, int slot)
 {
 	int rv = 0;
@@ -2012,6 +2014,7 @@ int md_bitmap_copy_from_slot(struct mddev *mddev, int slot,
 	md_bitmap_unplug(mddev->bitmap);
 	*low = lo;
 	*high = hi;
+	md_bitmap_free(bitmap);
 
 	return rv;
 }
@@ -2615,4 +2618,3 @@ struct attribute_group md_bitmap_group = {
 	.name = "bitmap",
 	.attrs = md_bitmap_attrs,
 };
-
diff --git a/drivers/md/md-cluster.c b/drivers/md/md-cluster.c
index 0580b51a156a..4aaf4820b6f6 100644
--- a/drivers/md/md-cluster.c
+++ b/drivers/md/md-cluster.c
@@ -1166,6 +1166,7 @@ static int resize_bitmaps(struct mddev *mddev, sector_t newsize, sector_t oldsiz
 			 * can't resize bitmap
 			 */
 			goto out;
+		md_bitmap_free(bitmap);
 	}
 
 	return 0;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index de8419b7ae98..98bac4f304ae 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -8582,6 +8582,26 @@ void md_write_end(struct mddev *mddev)
 
 EXPORT_SYMBOL(md_write_end);
 
+/* This is used by raid0 and raid10 */
+void md_submit_discard_bio(struct mddev *mddev, struct md_rdev *rdev,
+			struct bio *bio, sector_t start, sector_t size)
+{
+	struct bio *discard_bio = NULL;
+
+	if (__blkdev_issue_discard(rdev->bdev, start, size,
+		GFP_NOIO, 0, &discard_bio) || !discard_bio)
+		return;
+
+	bio_chain(discard_bio, bio);
+	bio_clone_blkg_association(discard_bio, bio);
+	if (mddev->gendisk)
+		trace_block_bio_remap(bdev_get_queue(rdev->bdev),
+			discard_bio, disk_devt(mddev->gendisk),
+			bio->bi_iter.bi_sector);
+	submit_bio_noacct(discard_bio);
+}
+EXPORT_SYMBOL(md_submit_discard_bio);
+
 /* md_allow_write(mddev)
  * Calling this ensures that the array is marked 'active' so that writes
  * may proceed without blocking.  It is important to call this before
@@ -9544,7 +9564,7 @@ static int __init md_init(void)
 		goto err_misc_wq;
 
 	md_rdev_misc_wq = alloc_workqueue("md_rdev_misc", 0, 0);
-	if (!md_misc_wq)
+	if (!md_rdev_misc_wq)
 		goto err_rdev_misc_wq;
 
 	if ((ret = register_blkdev(MD_MAJOR, "md")) < 0)
diff --git a/drivers/md/md.h b/drivers/md/md.h
index 2175a5ac4f7c..ccfb69868c2e 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -713,6 +713,8 @@ extern void md_write_end(struct mddev *mddev);
 extern void md_done_sync(struct mddev *mddev, int blocks, int ok);
 extern void md_error(struct mddev *mddev, struct md_rdev *rdev);
 extern void md_finish_reshape(struct mddev *mddev);
+extern void md_submit_discard_bio(struct mddev *mddev, struct md_rdev *rdev,
+			struct bio *bio, sector_t start, sector_t size);
 
 extern bool __must_check md_flush_request(struct mddev *mddev, struct bio *bio);
 extern void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index aa2d72791768..6f44177593a5 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -426,23 +426,6 @@ static void raid0_free(struct mddev *mddev, void *priv)
 	kfree(conf);
 }
 
-/*
- * Is io distribute over 1 or more chunks ?
-*/
-static inline int is_io_in_chunk_boundary(struct mddev *mddev,
-			unsigned int chunk_sects, struct bio *bio)
-{
-	if (likely(is_power_of_2(chunk_sects))) {
-		return chunk_sects >=
-			((bio->bi_iter.bi_sector & (chunk_sects-1))
-					+ bio_sectors(bio));
-	} else{
-		sector_t sector = bio->bi_iter.bi_sector;
-		return chunk_sects >= (sector_div(sector, chunk_sects)
-						+ bio_sectors(bio));
-	}
-}
-
 static void raid0_handle_discard(struct mddev *mddev, struct bio *bio)
 {
 	struct r0conf *conf = mddev->private;
@@ -494,7 +477,6 @@ static void raid0_handle_discard(struct mddev *mddev, struct bio *bio)
 
 	for (disk = 0; disk < zone->nb_dev; disk++) {
 		sector_t dev_start, dev_end;
-		struct bio *discard_bio = NULL;
 		struct md_rdev *rdev;
 
 		if (disk < start_disk_index)
@@ -517,18 +499,9 @@ static void raid0_handle_discard(struct mddev *mddev, struct bio *bio)
 
 		rdev = conf->devlist[(zone - conf->strip_zone) *
 			conf->strip_zone[0].nb_dev + disk];
-		if (__blkdev_issue_discard(rdev->bdev,
+		md_submit_discard_bio(mddev, rdev, bio,
 			dev_start + zone->dev_start + rdev->data_offset,
-			dev_end - dev_start, GFP_NOIO, 0, &discard_bio) ||
-		    !discard_bio)
-			continue;
-		bio_chain(discard_bio, bio);
-		bio_clone_blkg_association(discard_bio, bio);
-		if (mddev->gendisk)
-			trace_block_bio_remap(bdev_get_queue(rdev->bdev),
-				discard_bio, disk_devt(mddev->gendisk),
-				bio->bi_iter.bi_sector);
-		submit_bio_noacct(discard_bio);
+			dev_end - dev_start);
 	}
 	bio_endio(bio);
 }
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 5d1bdee313ec..b7bca6703df8 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -91,7 +91,7 @@ static inline struct r10bio *get_resync_r10bio(struct bio *bio)
 static void * r10bio_pool_alloc(gfp_t gfp_flags, void *data)
 {
 	struct r10conf *conf = data;
-	int size = offsetof(struct r10bio, devs[conf->copies]);
+	int size = offsetof(struct r10bio, devs[conf->geo.raid_disks]);
 
 	/* allocate a r10bio with room for raid_disks entries in the
 	 * bios array */
@@ -238,7 +238,7 @@ static void put_all_bios(struct r10conf *conf, struct r10bio *r10_bio)
 {
 	int i;
 
-	for (i = 0; i < conf->copies; i++) {
+	for (i = 0; i < conf->geo.raid_disks; i++) {
 		struct bio **bio = & r10_bio->devs[i].bio;
 		if (!BIO_SPECIAL(*bio))
 			bio_put(*bio);
@@ -327,7 +327,7 @@ static int find_bio_disk(struct r10conf *conf, struct r10bio *r10_bio,
 	int slot;
 	int repl = 0;
 
-	for (slot = 0; slot < conf->copies; slot++) {
+	for (slot = 0; slot < conf->geo.raid_disks; slot++) {
 		if (r10_bio->devs[slot].bio == bio)
 			break;
 		if (r10_bio->devs[slot].repl_bio == bio) {
@@ -336,7 +336,6 @@ static int find_bio_disk(struct r10conf *conf, struct r10bio *r10_bio,
 		}
 	}
 
-	BUG_ON(slot == conf->copies);
 	update_head_pos(slot, r10_bio);
 
 	if (slotp)
@@ -1276,12 +1275,75 @@ static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio,
 	}
 }
 
+static void wait_blocked_dev(struct mddev *mddev, struct r10bio *r10_bio)
+{
+	int i;
+	struct r10conf *conf = mddev->private;
+	struct md_rdev *blocked_rdev;
+
+retry_wait:
+	blocked_rdev = NULL;
+	rcu_read_lock();
+	for (i = 0; i < conf->copies; i++) {
+		struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev);
+		struct md_rdev *rrdev = rcu_dereference(
+			conf->mirrors[i].replacement);
+		if (rdev == rrdev)
+			rrdev = NULL;
+		if (rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
+			atomic_inc(&rdev->nr_pending);
+			blocked_rdev = rdev;
+			break;
+		}
+		if (rrdev && unlikely(test_bit(Blocked, &rrdev->flags))) {
+			atomic_inc(&rrdev->nr_pending);
+			blocked_rdev = rrdev;
+			break;
+		}
+
+		if (rdev && test_bit(WriteErrorSeen, &rdev->flags)) {
+			sector_t first_bad;
+			sector_t dev_sector = r10_bio->devs[i].addr;
+			int bad_sectors;
+			int is_bad;
+
+			/* Discard request doesn't care the write result
+			 * so it doesn't need to wait blocked disk here.
+			 */
+			if (!r10_bio->sectors)
+				continue;
+
+			is_bad = is_badblock(rdev, dev_sector, r10_bio->sectors,
+					     &first_bad, &bad_sectors);
+			if (is_bad < 0) {
+				/* Mustn't write here until the bad block
+				 * is acknowledged
+				 */
+				atomic_inc(&rdev->nr_pending);
+				set_bit(BlockedBadBlocks, &rdev->flags);
+				blocked_rdev = rdev;
+				break;
+			}
+		}
+	}
+	rcu_read_unlock();
+
+	if (unlikely(blocked_rdev)) {
+		/* Have to wait for this device to get unblocked, then retry */
+		allow_barrier(conf);
+		raid10_log(conf->mddev, "%s wait rdev %d blocked",
+				__func__, blocked_rdev->raid_disk);
+		md_wait_for_blocked_rdev(blocked_rdev, mddev);
+		wait_barrier(conf);
+		goto retry_wait;
+	}
+}
+
 static void raid10_write_request(struct mddev *mddev, struct bio *bio,
 				 struct r10bio *r10_bio)
 {
 	struct r10conf *conf = mddev->private;
 	int i;
-	struct md_rdev *blocked_rdev;
 	sector_t sectors;
 	int max_sectors;
 
@@ -1339,8 +1401,9 @@ static void raid10_write_request(struct mddev *mddev, struct bio *bio,
 
 	r10_bio->read_slot = -1; /* make sure repl_bio gets freed */
 	raid10_find_phys(conf, r10_bio);
-retry_write:
-	blocked_rdev = NULL;
+
+	wait_blocked_dev(mddev, r10_bio);
+
 	rcu_read_lock();
 	max_sectors = r10_bio->sectors;
 
@@ -1351,16 +1414,6 @@ retry_write:
 			conf->mirrors[d].replacement);
 		if (rdev == rrdev)
 			rrdev = NULL;
-		if (rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
-			atomic_inc(&rdev->nr_pending);
-			blocked_rdev = rdev;
-			break;
-		}
-		if (rrdev && unlikely(test_bit(Blocked, &rrdev->flags))) {
-			atomic_inc(&rrdev->nr_pending);
-			blocked_rdev = rrdev;
-			break;
-		}
 		if (rdev && (test_bit(Faulty, &rdev->flags)))
 			rdev = NULL;
 		if (rrdev && (test_bit(Faulty, &rrdev->flags)))
@@ -1381,15 +1434,6 @@ retry_write:
 
 			is_bad = is_badblock(rdev, dev_sector, max_sectors,
 					     &first_bad, &bad_sectors);
-			if (is_bad < 0) {
-				/* Mustn't write here until the bad block
-				 * is acknowledged
-				 */
-				atomic_inc(&rdev->nr_pending);
-				set_bit(BlockedBadBlocks, &rdev->flags);
-				blocked_rdev = rdev;
-				break;
-			}
 			if (is_bad && first_bad <= dev_sector) {
 				/* Cannot write here at all */
 				bad_sectors -= (dev_sector - first_bad);
@@ -1425,35 +1469,6 @@ retry_write:
 	}
 	rcu_read_unlock();
 
-	if (unlikely(blocked_rdev)) {
-		/* Have to wait for this device to get unblocked, then retry */
-		int j;
-		int d;
-
-		for (j = 0; j < i; j++) {
-			if (r10_bio->devs[j].bio) {
-				d = r10_bio->devs[j].devnum;
-				rdev_dec_pending(conf->mirrors[d].rdev, mddev);
-			}
-			if (r10_bio->devs[j].repl_bio) {
-				struct md_rdev *rdev;
-				d = r10_bio->devs[j].devnum;
-				rdev = conf->mirrors[d].replacement;
-				if (!rdev) {
-					/* Race with remove_disk */
-					smp_mb();
-					rdev = conf->mirrors[d].rdev;
-				}
-				rdev_dec_pending(rdev, mddev);
-			}
-		}
-		allow_barrier(conf);
-		raid10_log(conf->mddev, "wait rdev %d blocked", blocked_rdev->raid_disk);
-		md_wait_for_blocked_rdev(blocked_rdev, mddev);
-		wait_barrier(conf);
-		goto retry_write;
-	}
-
 	if (max_sectors < r10_bio->sectors)
 		r10_bio->sectors = max_sectors;
 
@@ -1493,7 +1508,7 @@ static void __make_request(struct mddev *mddev, struct bio *bio, int sectors)
 	r10_bio->mddev = mddev;
 	r10_bio->sector = bio->bi_iter.bi_sector;
 	r10_bio->state = 0;
-	memset(r10_bio->devs, 0, sizeof(r10_bio->devs[0]) * conf->copies);
+	memset(r10_bio->devs, 0, sizeof(r10_bio->devs[0]) * conf->geo.raid_disks);
 
 	if (bio_data_dir(bio) == READ)
 		raid10_read_request(mddev, bio, r10_bio);
@@ -1501,6 +1516,296 @@ static void __make_request(struct mddev *mddev, struct bio *bio, int sectors)
 		raid10_write_request(mddev, bio, r10_bio);
 }
 
+static struct bio *raid10_split_bio(struct r10conf *conf,
+			struct bio *bio, sector_t sectors, bool want_first)
+{
+	struct bio *split;
+
+	split = bio_split(bio, sectors,	GFP_NOIO, &conf->bio_split);
+	bio_chain(split, bio);
+	allow_barrier(conf);
+	if (want_first) {
+		submit_bio_noacct(bio);
+		bio = split;
+	} else
+		submit_bio_noacct(split);
+	wait_barrier(conf);
+
+	return bio;
+}
+
+static void raid_end_discard_bio(struct r10bio *r10bio)
+{
+	struct r10conf *conf = r10bio->mddev->private;
+	struct r10bio *first_r10bio;
+
+	while (atomic_dec_and_test(&r10bio->remaining)) {
+
+		allow_barrier(conf);
+
+		if (!test_bit(R10BIO_Discard, &r10bio->state)) {
+			first_r10bio = (struct r10bio *)r10bio->master_bio;
+			free_r10bio(r10bio);
+			r10bio = first_r10bio;
+		} else {
+			md_write_end(r10bio->mddev);
+			bio_endio(r10bio->master_bio);
+			free_r10bio(r10bio);
+			break;
+		}
+	}
+}
+
+static void raid10_end_discard_request(struct bio *bio)
+{
+	struct r10bio *r10_bio = bio->bi_private;
+	struct r10conf *conf = r10_bio->mddev->private;
+	struct md_rdev *rdev = NULL;
+	int dev;
+	int slot, repl;
+
+	/*
+	 * We don't care the return value of discard bio
+	 */
+	if (!test_bit(R10BIO_Uptodate, &r10_bio->state))
+		set_bit(R10BIO_Uptodate, &r10_bio->state);
+
+	dev = find_bio_disk(conf, r10_bio, bio, &slot, &repl);
+	if (repl)
+		rdev = conf->mirrors[dev].replacement;
+	if (!rdev) {
+		/* raid10_remove_disk uses smp_mb to make sure rdev is set to
+		 * replacement before setting replacement to NULL. It can read
+		 * rdev first without barrier protect even replacment is NULL
+		 */
+		smp_rmb();
+		rdev = conf->mirrors[dev].rdev;
+	}
+
+	raid_end_discard_bio(r10_bio);
+	rdev_dec_pending(rdev, conf->mddev);
+}
+
+/* There are some limitations to handle discard bio
+ * 1st, the discard size is bigger than stripe_size*2.
+ * 2st, if the discard bio spans reshape progress, we use the old way to
+ * handle discard bio
+ */
+static int raid10_handle_discard(struct mddev *mddev, struct bio *bio)
+{
+	struct r10conf *conf = mddev->private;
+	struct geom *geo = &conf->geo;
+	struct r10bio *r10_bio, *first_r10bio;
+	int far_copies = geo->far_copies;
+	bool first_copy = true;
+
+	int disk;
+	sector_t chunk;
+	unsigned int stripe_size;
+	sector_t split_size;
+
+	sector_t bio_start, bio_end;
+	sector_t first_stripe_index, last_stripe_index;
+	sector_t start_disk_offset;
+	unsigned int start_disk_index;
+	sector_t end_disk_offset;
+	unsigned int end_disk_index;
+	unsigned int remainder;
+
+	if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
+		return -EAGAIN;
+
+	wait_barrier(conf);
+
+	/* Check reshape again to avoid reshape happens after checking
+	 * MD_RECOVERY_RESHAPE and before wait_barrier
+	 */
+	if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
+		goto out;
+
+	stripe_size = geo->raid_disks << geo->chunk_shift;
+	bio_start = bio->bi_iter.bi_sector;
+	bio_end = bio_end_sector(bio);
+
+	/* Maybe one discard bio is smaller than strip size or across one stripe
+	 * and discard region is larger than one stripe size. For far offset layout,
+	 * if the discard region is not aligned with stripe size, there is hole
+	 * when we submit discard bio to member disk. For simplicity, we only
+	 * handle discard bio which discard region is bigger than stripe_size*2
+	 */
+	if (bio_sectors(bio) < stripe_size*2)
+		goto out;
+
+	/* For far and far offset layout, if bio is not aligned with stripe size,
+	 * it splits the part that is not aligned with strip size.
+	 */
+	div_u64_rem(bio_start, stripe_size, &remainder);
+	if ((far_copies > 1) && remainder) {
+		split_size = stripe_size - remainder;
+		bio = raid10_split_bio(conf, bio, split_size, false);
+	}
+	div_u64_rem(bio_end, stripe_size, &remainder);
+	if ((far_copies > 1) && remainder) {
+		split_size = bio_sectors(bio) - remainder;
+		bio = raid10_split_bio(conf, bio, split_size, true);
+	}
+
+	bio_start = bio->bi_iter.bi_sector;
+	bio_end = bio_end_sector(bio);
+
+	/* raid10 uses chunk as the unit to store data. It's similar like raid0.
+	 * One stripe contains the chunks from all member disk (one chunk from
+	 * one disk at the same HBA address). For layout detail, see 'man md 4'
+	 */
+	chunk = bio_start >> geo->chunk_shift;
+	chunk *= geo->near_copies;
+	first_stripe_index = chunk;
+	start_disk_index = sector_div(first_stripe_index, geo->raid_disks);
+	if (geo->far_offset)
+		first_stripe_index *= geo->far_copies;
+	start_disk_offset = (bio_start & geo->chunk_mask) +
+				(first_stripe_index << geo->chunk_shift);
+
+	chunk = bio_end >> geo->chunk_shift;
+	chunk *= geo->near_copies;
+	last_stripe_index = chunk;
+	end_disk_index = sector_div(last_stripe_index, geo->raid_disks);
+	if (geo->far_offset)
+		last_stripe_index *= geo->far_copies;
+	end_disk_offset = (bio_end & geo->chunk_mask) +
+				(last_stripe_index << geo->chunk_shift);
+
+retry_discard:
+	r10_bio = mempool_alloc(&conf->r10bio_pool, GFP_NOIO);
+	r10_bio->mddev = mddev;
+	r10_bio->state = 0;
+	r10_bio->sectors = 0;
+	memset(r10_bio->devs, 0, sizeof(r10_bio->devs[0]) * geo->raid_disks);
+	wait_blocked_dev(mddev, r10_bio);
+
+	/* For far layout it needs more than one r10bio to cover all regions.
+	 * Inspired by raid10_sync_request, we can use the first r10bio->master_bio
+	 * to record the discard bio. Other r10bio->master_bio record the first
+	 * r10bio. The first r10bio only release after all other r10bios finish.
+	 * The discard bio returns only first r10bio finishes
+	 */
+	if (first_copy) {
+		r10_bio->master_bio = bio;
+		set_bit(R10BIO_Discard, &r10_bio->state);
+		first_copy = false;
+		first_r10bio = r10_bio;
+	} else
+		r10_bio->master_bio = (struct bio *)first_r10bio;
+
+	rcu_read_lock();
+	for (disk = 0; disk < geo->raid_disks; disk++) {
+		struct md_rdev *rdev = rcu_dereference(conf->mirrors[disk].rdev);
+		struct md_rdev *rrdev = rcu_dereference(
+			conf->mirrors[disk].replacement);
+
+		r10_bio->devs[disk].bio = NULL;
+		r10_bio->devs[disk].repl_bio = NULL;
+
+		if (rdev && (test_bit(Faulty, &rdev->flags)))
+			rdev = NULL;
+		if (rrdev && (test_bit(Faulty, &rrdev->flags)))
+			rrdev = NULL;
+		if (!rdev && !rrdev)
+			continue;
+
+		if (rdev) {
+			r10_bio->devs[disk].bio = bio;
+			atomic_inc(&rdev->nr_pending);
+		}
+		if (rrdev) {
+			r10_bio->devs[disk].repl_bio = bio;
+			atomic_inc(&rrdev->nr_pending);
+		}
+	}
+	rcu_read_unlock();
+
+	atomic_set(&r10_bio->remaining, 1);
+	for (disk = 0; disk < geo->raid_disks; disk++) {
+		sector_t dev_start, dev_end;
+		struct bio *mbio, *rbio = NULL;
+		struct md_rdev *rdev = rcu_dereference(conf->mirrors[disk].rdev);
+		struct md_rdev *rrdev = rcu_dereference(
+			conf->mirrors[disk].replacement);
+
+		/*
+		 * Now start to calculate the start and end address for each disk.
+		 * The space between dev_start and dev_end is the discard region.
+		 *
+		 * For dev_start, it needs to consider three conditions:
+		 * 1st, the disk is before start_disk, you can imagine the disk in
+		 * the next stripe. So the dev_start is the start address of next
+		 * stripe.
+		 * 2st, the disk is after start_disk, it means the disk is at the
+		 * same stripe of first disk
+		 * 3st, the first disk itself, we can use start_disk_offset directly
+		 */
+		if (disk < start_disk_index)
+			dev_start = (first_stripe_index + 1) * mddev->chunk_sectors;
+		else if (disk > start_disk_index)
+			dev_start = first_stripe_index * mddev->chunk_sectors;
+		else
+			dev_start = start_disk_offset;
+
+		if (disk < end_disk_index)
+			dev_end = (last_stripe_index + 1) * mddev->chunk_sectors;
+		else if (disk > end_disk_index)
+			dev_end = last_stripe_index * mddev->chunk_sectors;
+		else
+			dev_end = end_disk_offset;
+
+		/* It only handles discard bio which size is >= stripe size, so
+		 * dev_end > dev_start all the time
+		 */
+		if (r10_bio->devs[disk].bio) {
+			mbio = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set);
+			mbio->bi_end_io = raid10_end_discard_request;
+			mbio->bi_private = r10_bio;
+			r10_bio->devs[disk].bio = mbio;
+			r10_bio->devs[disk].devnum = disk;
+			atomic_inc(&r10_bio->remaining);
+			md_submit_discard_bio(mddev, rdev, mbio,
+					dev_start + choose_data_offset(r10_bio, rdev),
+					dev_end - dev_start);
+			bio_endio(mbio);
+		}
+		if (r10_bio->devs[disk].repl_bio) {
+			rbio = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set);
+			rbio->bi_end_io = raid10_end_discard_request;
+			rbio->bi_private = r10_bio;
+			r10_bio->devs[disk].repl_bio = rbio;
+			r10_bio->devs[disk].devnum = disk;
+			atomic_inc(&r10_bio->remaining);
+			md_submit_discard_bio(mddev, rrdev, rbio,
+					dev_start + choose_data_offset(r10_bio, rrdev),
+					dev_end - dev_start);
+			bio_endio(rbio);
+		}
+	}
+
+	if (!geo->far_offset && --far_copies) {
+		first_stripe_index += geo->stride >> geo->chunk_shift;
+		start_disk_offset += geo->stride;
+		last_stripe_index += geo->stride >> geo->chunk_shift;
+		end_disk_offset += geo->stride;
+		atomic_inc(&first_r10bio->remaining);
+		raid_end_discard_bio(r10_bio);
+		wait_barrier(conf);
+		goto retry_discard;
+	}
+
+	raid_end_discard_bio(r10_bio);
+
+	return 0;
+out:
+	allow_barrier(conf);
+	return -EAGAIN;
+}
+
 static bool raid10_make_request(struct mddev *mddev, struct bio *bio)
 {
 	struct r10conf *conf = mddev->private;
@@ -1515,6 +1820,10 @@ static bool raid10_make_request(struct mddev *mddev, struct bio *bio)
 	if (!md_write_start(mddev, bio))
 		return false;
 
+	if (unlikely(bio_op(bio) == REQ_OP_DISCARD))
+		if (!raid10_handle_discard(mddev, bio))
+			return true;
+
 	/*
 	 * If this request crosses a chunk boundary, we need to split
 	 * it.
@@ -3754,7 +4063,7 @@ static int raid10_run(struct mddev *mddev)
 
 	if (mddev->queue) {
 		blk_queue_max_discard_sectors(mddev->queue,
-					      mddev->chunk_sectors);
+					      UINT_MAX);
 		blk_queue_max_write_same_sectors(mddev->queue, 0);
 		blk_queue_max_write_zeroes_sectors(mddev->queue, 0);
 		blk_queue_io_min(mddev->queue, mddev->chunk_sectors << 9);
@@ -4458,8 +4767,8 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
 		last = conf->reshape_progress - 1;
 		sector_nr = last & ~(sector_t)(conf->geo.chunk_mask
 					       & conf->prev.chunk_mask);
-		if (sector_nr + RESYNC_BLOCK_SIZE/512 < last)
-			sector_nr = last + 1 - RESYNC_BLOCK_SIZE/512;
+		if (sector_nr + RESYNC_SECTORS < last)
+			sector_nr = last + 1 - RESYNC_SECTORS;
 	} else {
 		/* 'next' is after the last device address that we
 		 * might write to for this chunk in the new layout
@@ -4481,8 +4790,8 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
 		last  = sector_nr | (conf->geo.chunk_mask
 				     & conf->prev.chunk_mask);
 
-		if (sector_nr + RESYNC_BLOCK_SIZE/512 <= last)
-			last = sector_nr + RESYNC_BLOCK_SIZE/512 - 1;
+		if (sector_nr + RESYNC_SECTORS <= last)
+			last = sector_nr + RESYNC_SECTORS - 1;
 	}
 
 	if (need_flush ||
diff --git a/drivers/md/raid10.h b/drivers/md/raid10.h
index 79cd2b7d3128..1461fd55311b 100644
--- a/drivers/md/raid10.h
+++ b/drivers/md/raid10.h
@@ -179,5 +179,6 @@ enum r10bio_state {
 	R10BIO_Previous,
 /* failfast devices did receive failfast requests. */
 	R10BIO_FailFast,
+	R10BIO_Discard,
 };
 #endif
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index d589d26c86ea..39343479ac2a 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -448,13 +448,74 @@ out:
 	return sh;
 }
 
-static void shrink_buffers(struct stripe_head *sh)
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+static void free_stripe_pages(struct stripe_head *sh)
+{
+	int i;
+	struct page *p;
+
+	/* Have not allocate page pool */
+	if (!sh->pages)
+		return;
+
+	for (i = 0; i < sh->nr_pages; i++) {
+		p = sh->pages[i];
+		if (p)
+			put_page(p);
+		sh->pages[i] = NULL;
+	}
+}
+
+static int alloc_stripe_pages(struct stripe_head *sh, gfp_t gfp)
 {
+	int i;
 	struct page *p;
+
+	for (i = 0; i < sh->nr_pages; i++) {
+		/* The page have allocated. */
+		if (sh->pages[i])
+			continue;
+
+		p = alloc_page(gfp);
+		if (!p) {
+			free_stripe_pages(sh);
+			return -ENOMEM;
+		}
+		sh->pages[i] = p;
+	}
+	return 0;
+}
+
+static int
+init_stripe_shared_pages(struct stripe_head *sh, struct r5conf *conf, int disks)
+{
+	int nr_pages, cnt;
+
+	if (sh->pages)
+		return 0;
+
+	/* Each of the sh->dev[i] need one conf->stripe_size */
+	cnt = PAGE_SIZE / conf->stripe_size;
+	nr_pages = (disks + cnt - 1) / cnt;
+
+	sh->pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL);
+	if (!sh->pages)
+		return -ENOMEM;
+	sh->nr_pages = nr_pages;
+	sh->stripes_per_page = cnt;
+	return 0;
+}
+#endif
+
+static void shrink_buffers(struct stripe_head *sh)
+{
 	int i;
 	int num = sh->raid_conf->pool_size;
 
+#if PAGE_SIZE == DEFAULT_STRIPE_SIZE
 	for (i = 0; i < num ; i++) {
+		struct page *p;
+
 		WARN_ON(sh->dev[i].page != sh->dev[i].orig_page);
 		p = sh->dev[i].page;
 		if (!p)
@@ -462,6 +523,11 @@ static void shrink_buffers(struct stripe_head *sh)
 		sh->dev[i].page = NULL;
 		put_page(p);
 	}
+#else
+	for (i = 0; i < num; i++)
+		sh->dev[i].page = NULL;
+	free_stripe_pages(sh); /* Free pages */
+#endif
 }
 
 static int grow_buffers(struct stripe_head *sh, gfp_t gfp)
@@ -469,6 +535,7 @@ static int grow_buffers(struct stripe_head *sh, gfp_t gfp)
 	int i;
 	int num = sh->raid_conf->pool_size;
 
+#if PAGE_SIZE == DEFAULT_STRIPE_SIZE
 	for (i = 0; i < num; i++) {
 		struct page *page;
 
@@ -477,8 +544,18 @@ static int grow_buffers(struct stripe_head *sh, gfp_t gfp)
 		}
 		sh->dev[i].page = page;
 		sh->dev[i].orig_page = page;
+		sh->dev[i].offset = 0;
 	}
+#else
+	if (alloc_stripe_pages(sh, gfp))
+		return -ENOMEM;
 
+	for (i = 0; i < num; i++) {
+		sh->dev[i].page = raid5_get_dev_page(sh, i);
+		sh->dev[i].orig_page = sh->dev[i].page;
+		sh->dev[i].offset = raid5_get_page_offset(sh, i);
+	}
+#endif
 	return 0;
 }
 
@@ -1130,7 +1207,7 @@ again:
 				sh->dev[i].vec.bv_page = sh->dev[i].page;
 			bi->bi_vcnt = 1;
 			bi->bi_io_vec[0].bv_len = RAID5_STRIPE_SIZE(conf);
-			bi->bi_io_vec[0].bv_offset = 0;
+			bi->bi_io_vec[0].bv_offset = sh->dev[i].offset;
 			bi->bi_iter.bi_size = RAID5_STRIPE_SIZE(conf);
 			bi->bi_write_hint = sh->dev[i].write_hint;
 			if (!rrdev)
@@ -1184,7 +1261,7 @@ again:
 			sh->dev[i].rvec.bv_page = sh->dev[i].page;
 			rbi->bi_vcnt = 1;
 			rbi->bi_io_vec[0].bv_len = RAID5_STRIPE_SIZE(conf);
-			rbi->bi_io_vec[0].bv_offset = 0;
+			rbi->bi_io_vec[0].bv_offset = sh->dev[i].offset;
 			rbi->bi_iter.bi_size = RAID5_STRIPE_SIZE(conf);
 			rbi->bi_write_hint = sh->dev[i].write_hint;
 			sh->dev[i].write_hint = RWH_WRITE_LIFE_NOT_SET;
@@ -1226,7 +1303,7 @@ again:
 
 static struct dma_async_tx_descriptor *
 async_copy_data(int frombio, struct bio *bio, struct page **page,
-	sector_t sector, struct dma_async_tx_descriptor *tx,
+	unsigned int poff, sector_t sector, struct dma_async_tx_descriptor *tx,
 	struct stripe_head *sh, int no_skipcopy)
 {
 	struct bio_vec bvl;
@@ -1272,11 +1349,11 @@ async_copy_data(int frombio, struct bio *bio, struct page **page,
 				    !no_skipcopy)
 					*page = bio_page;
 				else
-					tx = async_memcpy(*page, bio_page, page_offset,
+					tx = async_memcpy(*page, bio_page, page_offset + poff,
 						  b_offset, clen, &submit);
 			} else
 				tx = async_memcpy(bio_page, *page, b_offset,
-						  page_offset, clen, &submit);
+						  page_offset + poff, clen, &submit);
 		}
 		/* chain the operations */
 		submit.depend_tx = tx;
@@ -1349,6 +1426,7 @@ static void ops_run_biofill(struct stripe_head *sh)
 			while (rbi && rbi->bi_iter.bi_sector <
 				dev->sector + RAID5_STRIPE_SECTORS(conf)) {
 				tx = async_copy_data(0, rbi, &dev->page,
+						     dev->offset,
 						     dev->sector, tx, sh, 0);
 				rbi = r5_next_bio(conf, rbi, dev->sector);
 			}
@@ -1404,14 +1482,25 @@ static addr_conv_t *to_addr_conv(struct stripe_head *sh,
 	return (void *) (to_addr_page(percpu, i) + sh->disks + 2);
 }
 
+/*
+ * Return a pointer to record offset address.
+ */
+static unsigned int *
+to_addr_offs(struct stripe_head *sh, struct raid5_percpu *percpu)
+{
+	return (unsigned int *) (to_addr_conv(sh, percpu, 0) + sh->disks + 2);
+}
+
 static struct dma_async_tx_descriptor *
 ops_run_compute5(struct stripe_head *sh, struct raid5_percpu *percpu)
 {
 	int disks = sh->disks;
 	struct page **xor_srcs = to_addr_page(percpu, 0);
+	unsigned int *off_srcs = to_addr_offs(sh, percpu);
 	int target = sh->ops.target;
 	struct r5dev *tgt = &sh->dev[target];
 	struct page *xor_dest = tgt->page;
+	unsigned int off_dest = tgt->offset;
 	int count = 0;
 	struct dma_async_tx_descriptor *tx;
 	struct async_submit_ctl submit;
@@ -1423,19 +1512,22 @@ ops_run_compute5(struct stripe_head *sh, struct raid5_percpu *percpu)
 		__func__, (unsigned long long)sh->sector, target);
 	BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
 
-	for (i = disks; i--; )
-		if (i != target)
+	for (i = disks; i--; ) {
+		if (i != target) {
+			off_srcs[count] = sh->dev[i].offset;
 			xor_srcs[count++] = sh->dev[i].page;
+		}
+	}
 
 	atomic_inc(&sh->count);
 
 	init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST, NULL,
 			  ops_complete_compute, sh, to_addr_conv(sh, percpu, 0));
 	if (unlikely(count == 1))
-		tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0,
+		tx = async_memcpy(xor_dest, xor_srcs[0], off_dest, off_srcs[0],
 				RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 	else
-		tx = async_xor(xor_dest, xor_srcs, 0, count,
+		tx = async_xor_offs(xor_dest, off_dest, xor_srcs, off_srcs, count,
 				RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 
 	return tx;
@@ -1443,6 +1535,7 @@ ops_run_compute5(struct stripe_head *sh, struct raid5_percpu *percpu)
 
 /* set_syndrome_sources - populate source buffers for gen_syndrome
  * @srcs - (struct page *) array of size sh->disks
+ * @offs - (unsigned int) array of offset for each page
  * @sh - stripe_head to parse
  *
  * Populates srcs in proper layout order for the stripe and returns the
@@ -1451,6 +1544,7 @@ ops_run_compute5(struct stripe_head *sh, struct raid5_percpu *percpu)
  * is recorded in srcs[count+1]].
  */
 static int set_syndrome_sources(struct page **srcs,
+				unsigned int *offs,
 				struct stripe_head *sh,
 				int srctype)
 {
@@ -1481,6 +1575,12 @@ static int set_syndrome_sources(struct page **srcs,
 				srcs[slot] = sh->dev[i].orig_page;
 			else
 				srcs[slot] = sh->dev[i].page;
+			/*
+			 * For R5_InJournal, PAGE_SIZE must be 4KB and will
+			 * not shared page. In that case, dev[i].offset
+			 * is 0.
+			 */
+			offs[slot] = sh->dev[i].offset;
 		}
 		i = raid6_next_disk(i, disks);
 	} while (i != d0_idx);
@@ -1493,12 +1593,14 @@ ops_run_compute6_1(struct stripe_head *sh, struct raid5_percpu *percpu)
 {
 	int disks = sh->disks;
 	struct page **blocks = to_addr_page(percpu, 0);
+	unsigned int *offs = to_addr_offs(sh, percpu);
 	int target;
 	int qd_idx = sh->qd_idx;
 	struct dma_async_tx_descriptor *tx;
 	struct async_submit_ctl submit;
 	struct r5dev *tgt;
 	struct page *dest;
+	unsigned int dest_off;
 	int i;
 	int count;
 
@@ -1517,17 +1619,18 @@ ops_run_compute6_1(struct stripe_head *sh, struct raid5_percpu *percpu)
 	tgt = &sh->dev[target];
 	BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
 	dest = tgt->page;
+	dest_off = tgt->offset;
 
 	atomic_inc(&sh->count);
 
 	if (target == qd_idx) {
-		count = set_syndrome_sources(blocks, sh, SYNDROME_SRC_ALL);
+		count = set_syndrome_sources(blocks, offs, sh, SYNDROME_SRC_ALL);
 		blocks[count] = NULL; /* regenerating p is not necessary */
 		BUG_ON(blocks[count+1] != dest); /* q should already be set */
 		init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
 				  ops_complete_compute, sh,
 				  to_addr_conv(sh, percpu, 0));
-		tx = async_gen_syndrome(blocks, 0, count+2,
+		tx = async_gen_syndrome(blocks, offs, count+2,
 				RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 	} else {
 		/* Compute any data- or p-drive using XOR */
@@ -1535,13 +1638,14 @@ ops_run_compute6_1(struct stripe_head *sh, struct raid5_percpu *percpu)
 		for (i = disks; i-- ; ) {
 			if (i == target || i == qd_idx)
 				continue;
+			offs[count] = sh->dev[i].offset;
 			blocks[count++] = sh->dev[i].page;
 		}
 
 		init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST,
 				  NULL, ops_complete_compute, sh,
 				  to_addr_conv(sh, percpu, 0));
-		tx = async_xor(dest, blocks, 0, count,
+		tx = async_xor_offs(dest, dest_off, blocks, offs, count,
 				RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 	}
 
@@ -1561,6 +1665,7 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
 	struct r5dev *tgt2 = &sh->dev[target2];
 	struct dma_async_tx_descriptor *tx;
 	struct page **blocks = to_addr_page(percpu, 0);
+	unsigned int *offs = to_addr_offs(sh, percpu);
 	struct async_submit_ctl submit;
 
 	BUG_ON(sh->batch_head);
@@ -1573,13 +1678,16 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
 	/* we need to open-code set_syndrome_sources to handle the
 	 * slot number conversion for 'faila' and 'failb'
 	 */
-	for (i = 0; i < disks ; i++)
+	for (i = 0; i < disks ; i++) {
+		offs[i] = 0;
 		blocks[i] = NULL;
+	}
 	count = 0;
 	i = d0_idx;
 	do {
 		int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
 
+		offs[slot] = sh->dev[i].offset;
 		blocks[slot] = sh->dev[i].page;
 
 		if (i == target)
@@ -1604,11 +1712,12 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
 			init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
 					  ops_complete_compute, sh,
 					  to_addr_conv(sh, percpu, 0));
-			return async_gen_syndrome(blocks, 0, syndrome_disks+2,
+			return async_gen_syndrome(blocks, offs, syndrome_disks+2,
 						  RAID5_STRIPE_SIZE(sh->raid_conf),
 						  &submit);
 		} else {
 			struct page *dest;
+			unsigned int dest_off;
 			int data_target;
 			int qd_idx = sh->qd_idx;
 
@@ -1622,22 +1731,24 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
 			for (i = disks; i-- ; ) {
 				if (i == data_target || i == qd_idx)
 					continue;
+				offs[count] = sh->dev[i].offset;
 				blocks[count++] = sh->dev[i].page;
 			}
 			dest = sh->dev[data_target].page;
+			dest_off = sh->dev[data_target].offset;
 			init_async_submit(&submit,
 					  ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST,
 					  NULL, NULL, NULL,
 					  to_addr_conv(sh, percpu, 0));
-			tx = async_xor(dest, blocks, 0, count,
+			tx = async_xor_offs(dest, dest_off, blocks, offs, count,
 				       RAID5_STRIPE_SIZE(sh->raid_conf),
 				       &submit);
 
-			count = set_syndrome_sources(blocks, sh, SYNDROME_SRC_ALL);
+			count = set_syndrome_sources(blocks, offs, sh, SYNDROME_SRC_ALL);
 			init_async_submit(&submit, ASYNC_TX_FENCE, tx,
 					  ops_complete_compute, sh,
 					  to_addr_conv(sh, percpu, 0));
-			return async_gen_syndrome(blocks, 0, count+2,
+			return async_gen_syndrome(blocks, offs, count+2,
 						  RAID5_STRIPE_SIZE(sh->raid_conf),
 						  &submit);
 		}
@@ -1650,13 +1761,13 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
 			return async_raid6_datap_recov(syndrome_disks+2,
 						RAID5_STRIPE_SIZE(sh->raid_conf),
 						faila,
-						blocks, &submit);
+						blocks, offs, &submit);
 		} else {
 			/* We're missing D+D. */
 			return async_raid6_2data_recov(syndrome_disks+2,
 						RAID5_STRIPE_SIZE(sh->raid_conf),
 						faila, failb,
-						blocks, &submit);
+						blocks, offs, &submit);
 		}
 	}
 }
@@ -1682,10 +1793,12 @@ ops_run_prexor5(struct stripe_head *sh, struct raid5_percpu *percpu,
 {
 	int disks = sh->disks;
 	struct page **xor_srcs = to_addr_page(percpu, 0);
+	unsigned int *off_srcs = to_addr_offs(sh, percpu);
 	int count = 0, pd_idx = sh->pd_idx, i;
 	struct async_submit_ctl submit;
 
 	/* existing parity data subtracted */
+	unsigned int off_dest = off_srcs[count] = sh->dev[pd_idx].offset;
 	struct page *xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
 
 	BUG_ON(sh->batch_head);
@@ -1695,15 +1808,22 @@ ops_run_prexor5(struct stripe_head *sh, struct raid5_percpu *percpu,
 	for (i = disks; i--; ) {
 		struct r5dev *dev = &sh->dev[i];
 		/* Only process blocks that are known to be uptodate */
-		if (test_bit(R5_InJournal, &dev->flags))
+		if (test_bit(R5_InJournal, &dev->flags)) {
+			/*
+			 * For this case, PAGE_SIZE must be equal to 4KB and
+			 * page offset is zero.
+			 */
+			off_srcs[count] = dev->offset;
 			xor_srcs[count++] = dev->orig_page;
-		else if (test_bit(R5_Wantdrain, &dev->flags))
+		} else if (test_bit(R5_Wantdrain, &dev->flags)) {
+			off_srcs[count] = dev->offset;
 			xor_srcs[count++] = dev->page;
+		}
 	}
 
 	init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_DROP_DST, tx,
 			  ops_complete_prexor, sh, to_addr_conv(sh, percpu, 0));
-	tx = async_xor(xor_dest, xor_srcs, 0, count,
+	tx = async_xor_offs(xor_dest, off_dest, xor_srcs, off_srcs, count,
 			RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 
 	return tx;
@@ -1714,17 +1834,18 @@ ops_run_prexor6(struct stripe_head *sh, struct raid5_percpu *percpu,
 		struct dma_async_tx_descriptor *tx)
 {
 	struct page **blocks = to_addr_page(percpu, 0);
+	unsigned int *offs = to_addr_offs(sh, percpu);
 	int count;
 	struct async_submit_ctl submit;
 
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
 
-	count = set_syndrome_sources(blocks, sh, SYNDROME_SRC_WANT_DRAIN);
+	count = set_syndrome_sources(blocks, offs, sh, SYNDROME_SRC_WANT_DRAIN);
 
 	init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_PQ_XOR_DST, tx,
 			  ops_complete_prexor, sh, to_addr_conv(sh, percpu, 0));
-	tx = async_gen_syndrome(blocks, 0, count+2,
+	tx = async_gen_syndrome(blocks, offs, count+2,
 			RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 
 	return tx;
@@ -1775,6 +1896,7 @@ again:
 					set_bit(R5_Discard, &dev->flags);
 				else {
 					tx = async_copy_data(1, wbi, &dev->page,
+							     dev->offset,
 							     dev->sector, tx, sh,
 							     r5c_is_writeback(conf->log));
 					if (dev->page != dev->orig_page &&
@@ -1854,9 +1976,11 @@ ops_run_reconstruct5(struct stripe_head *sh, struct raid5_percpu *percpu,
 {
 	int disks = sh->disks;
 	struct page **xor_srcs;
+	unsigned int *off_srcs;
 	struct async_submit_ctl submit;
 	int count, pd_idx = sh->pd_idx, i;
 	struct page *xor_dest;
+	unsigned int off_dest;
 	int prexor = 0;
 	unsigned long flags;
 	int j = 0;
@@ -1881,24 +2005,31 @@ ops_run_reconstruct5(struct stripe_head *sh, struct raid5_percpu *percpu,
 again:
 	count = 0;
 	xor_srcs = to_addr_page(percpu, j);
+	off_srcs = to_addr_offs(sh, percpu);
 	/* check if prexor is active which means only process blocks
 	 * that are part of a read-modify-write (written)
 	 */
 	if (head_sh->reconstruct_state == reconstruct_state_prexor_drain_run) {
 		prexor = 1;
+		off_dest = off_srcs[count] = sh->dev[pd_idx].offset;
 		xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
 			if (head_sh->dev[i].written ||
-			    test_bit(R5_InJournal, &head_sh->dev[i].flags))
+			    test_bit(R5_InJournal, &head_sh->dev[i].flags)) {
+				off_srcs[count] = dev->offset;
 				xor_srcs[count++] = dev->page;
+			}
 		}
 	} else {
 		xor_dest = sh->dev[pd_idx].page;
+		off_dest = sh->dev[pd_idx].offset;
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
-			if (i != pd_idx)
+			if (i != pd_idx) {
+				off_srcs[count] = dev->offset;
 				xor_srcs[count++] = dev->page;
+			}
 		}
 	}
 
@@ -1924,10 +2055,10 @@ again:
 	}
 
 	if (unlikely(count == 1))
-		tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0,
+		tx = async_memcpy(xor_dest, xor_srcs[0], off_dest, off_srcs[0],
 				RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 	else
-		tx = async_xor(xor_dest, xor_srcs, 0, count,
+		tx = async_xor_offs(xor_dest, off_dest, xor_srcs, off_srcs, count,
 				RAID5_STRIPE_SIZE(sh->raid_conf), &submit);
 	if (!last_stripe) {
 		j++;
@@ -1943,6 +2074,7 @@ ops_run_reconstruct6(struct stripe_head *sh, struct raid5_percpu *percpu,
 {
 	struct async_submit_ctl submit;
 	struct page **blocks;
+	unsigned int *offs;
 	int count, i, j = 0;
 	struct stripe_head *head_sh = sh;
 	int last_stripe;
@@ -1967,6 +2099,7 @@ ops_run_reconstruct6(struct stripe_head *sh, struct raid5_percpu *percpu,
 
 again:
 	blocks = to_addr_page(percpu, j);
+	offs = to_addr_offs(sh, percpu);
 
 	if (sh->reconstruct_state == reconstruct_state_prexor_drain_run) {
 		synflags = SYNDROME_SRC_WRITTEN;
@@ -1976,7 +2109,7 @@ again:
 		txflags = ASYNC_TX_ACK;
 	}
 
-	count = set_syndrome_sources(blocks, sh, synflags);
+	count = set_syndrome_sources(blocks, offs, sh, synflags);
 	last_stripe = !head_sh->batch_head ||
 		list_first_entry(&sh->batch_list,
 				 struct stripe_head, batch_list) == head_sh;
@@ -1988,7 +2121,7 @@ again:
 	} else
 		init_async_submit(&submit, 0, tx, NULL, NULL,
 				  to_addr_conv(sh, percpu, j));
-	tx = async_gen_syndrome(blocks, 0, count+2,
+	tx = async_gen_syndrome(blocks, offs, count+2,
 			RAID5_STRIPE_SIZE(sh->raid_conf),  &submit);
 	if (!last_stripe) {
 		j++;
@@ -2016,7 +2149,9 @@ static void ops_run_check_p(struct stripe_head *sh, struct raid5_percpu *percpu)
 	int pd_idx = sh->pd_idx;
 	int qd_idx = sh->qd_idx;
 	struct page *xor_dest;
+	unsigned int off_dest;
 	struct page **xor_srcs = to_addr_page(percpu, 0);
+	unsigned int *off_srcs = to_addr_offs(sh, percpu);
 	struct dma_async_tx_descriptor *tx;
 	struct async_submit_ctl submit;
 	int count;
@@ -2028,16 +2163,19 @@ static void ops_run_check_p(struct stripe_head *sh, struct raid5_percpu *percpu)
 	BUG_ON(sh->batch_head);
 	count = 0;
 	xor_dest = sh->dev[pd_idx].page;
+	off_dest = sh->dev[pd_idx].offset;
+	off_srcs[count] = off_dest;
 	xor_srcs[count++] = xor_dest;
 	for (i = disks; i--; ) {
 		if (i == pd_idx || i == qd_idx)
 			continue;
+		off_srcs[count] = sh->dev[i].offset;
 		xor_srcs[count++] = sh->dev[i].page;
 	}
 
 	init_async_submit(&submit, 0, NULL, NULL, NULL,
 			  to_addr_conv(sh, percpu, 0));
-	tx = async_xor_val(xor_dest, xor_srcs, 0, count,
+	tx = async_xor_val_offs(xor_dest, off_dest, xor_srcs, off_srcs, count,
 			   RAID5_STRIPE_SIZE(sh->raid_conf),
 			   &sh->ops.zero_sum_result, &submit);
 
@@ -2049,6 +2187,7 @@ static void ops_run_check_p(struct stripe_head *sh, struct raid5_percpu *percpu)
 static void ops_run_check_pq(struct stripe_head *sh, struct raid5_percpu *percpu, int checkp)
 {
 	struct page **srcs = to_addr_page(percpu, 0);
+	unsigned int *offs = to_addr_offs(sh, percpu);
 	struct async_submit_ctl submit;
 	int count;
 
@@ -2056,16 +2195,16 @@ static void ops_run_check_pq(struct stripe_head *sh, struct raid5_percpu *percpu
 		(unsigned long long)sh->sector, checkp);
 
 	BUG_ON(sh->batch_head);
-	count = set_syndrome_sources(srcs, sh, SYNDROME_SRC_ALL);
+	count = set_syndrome_sources(srcs, offs, sh, SYNDROME_SRC_ALL);
 	if (!checkp)
 		srcs[count] = NULL;
 
 	atomic_inc(&sh->count);
 	init_async_submit(&submit, ASYNC_TX_ACK, NULL, ops_complete_check,
 			  sh, to_addr_conv(sh, percpu, 0));
-	async_syndrome_val(srcs, 0, count+2,
+	async_syndrome_val(srcs, offs, count+2,
 			   RAID5_STRIPE_SIZE(sh->raid_conf),
-			   &sh->ops.zero_sum_result, percpu->spare_page, &submit);
+			   &sh->ops.zero_sum_result, percpu->spare_page, 0, &submit);
 }
 
 static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
@@ -2142,6 +2281,9 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
 
 static void free_stripe(struct kmem_cache *sc, struct stripe_head *sh)
 {
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+	kfree(sh->pages);
+#endif
 	if (sh->ppl_page)
 		__free_page(sh->ppl_page);
 	kmem_cache_free(sc, sh);
@@ -2175,9 +2317,15 @@ static struct stripe_head *alloc_stripe(struct kmem_cache *sc, gfp_t gfp,
 			sh->ppl_page = alloc_page(gfp);
 			if (!sh->ppl_page) {
 				free_stripe(sc, sh);
-				sh = NULL;
+				return NULL;
 			}
 		}
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+		if (init_stripe_shared_pages(sh, conf, disks)) {
+			free_stripe(sc, sh);
+			return NULL;
+		}
+#endif
 	}
 	return sh;
 }
@@ -2253,8 +2401,9 @@ static int scribble_alloc(struct raid5_percpu *percpu,
 			  int num, int cnt)
 {
 	size_t obj_size =
-		sizeof(struct page *) * (num+2) +
-		sizeof(addr_conv_t) * (num+2);
+		sizeof(struct page *) * (num + 2) +
+		sizeof(addr_conv_t) * (num + 2) +
+		sizeof(unsigned int) * (num + 2);
 	void *scribble;
 
 	/*
@@ -2386,9 +2535,16 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 		osh = get_free_stripe(conf, hash);
 		unlock_device_hash_lock(conf, hash);
 
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+	for (i = 0; i < osh->nr_pages; i++) {
+		nsh->pages[i] = osh->pages[i];
+		osh->pages[i] = NULL;
+	}
+#endif
 		for(i=0; i<conf->pool_size; i++) {
 			nsh->dev[i].page = osh->dev[i].page;
 			nsh->dev[i].orig_page = osh->dev[i].page;
+			nsh->dev[i].offset = osh->dev[i].offset;
 		}
 		nsh->hash_lock_index = hash;
 		free_stripe(conf->slab_cache, osh);
@@ -2429,8 +2585,6 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 	} else
 		err = -ENOMEM;
 
-	mutex_unlock(&conf->cache_size_mutex);
-
 	conf->slab_cache = sc;
 	conf->active_name = 1-conf->active_name;
 
@@ -2439,20 +2593,41 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 		nsh = list_entry(newstripes.next, struct stripe_head, lru);
 		list_del_init(&nsh->lru);
 
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+		for (i = 0; i < nsh->nr_pages; i++) {
+			if (nsh->pages[i])
+				continue;
+			nsh->pages[i] = alloc_page(GFP_NOIO);
+			if (!nsh->pages[i])
+				err = -ENOMEM;
+		}
+
+		for (i = conf->raid_disks; i < newsize; i++) {
+			if (nsh->dev[i].page)
+				continue;
+			nsh->dev[i].page = raid5_get_dev_page(nsh, i);
+			nsh->dev[i].orig_page = nsh->dev[i].page;
+			nsh->dev[i].offset = raid5_get_page_offset(nsh, i);
+		}
+#else
 		for (i=conf->raid_disks; i < newsize; i++)
 			if (nsh->dev[i].page == NULL) {
 				struct page *p = alloc_page(GFP_NOIO);
 				nsh->dev[i].page = p;
 				nsh->dev[i].orig_page = p;
+				nsh->dev[i].offset = 0;
 				if (!p)
 					err = -ENOMEM;
 			}
+#endif
 		raid5_release_stripe(nsh);
 	}
 	/* critical section pass, GFP_NOIO no longer needed */
 
 	if (!err)
 		conf->pool_size = newsize;
+	mutex_unlock(&conf->cache_size_mutex);
+
 	return err;
 }
 
@@ -4369,7 +4544,8 @@ static void handle_stripe_expansion(struct r5conf *conf, struct stripe_head *sh)
 			/* place all the copies on one channel */
 			init_async_submit(&submit, 0, tx, NULL, NULL, NULL);
 			tx = async_memcpy(sh2->dev[dd_idx].page,
-					  sh->dev[i].page, 0, 0, RAID5_STRIPE_SIZE(conf),
+					  sh->dev[i].page, sh2->dev[dd_idx].offset,
+					  sh->dev[i].offset, RAID5_STRIPE_SIZE(conf),
 					  &submit);
 
 			set_bit(R5_Expanded, &sh2->dev[dd_idx].flags);
@@ -6506,6 +6682,7 @@ raid5_store_stripe_size(struct mddev  *mddev, const char *page, size_t len)
 	struct r5conf *conf;
 	unsigned long new;
 	int err;
+	int size;
 
 	if (len >= PAGE_SIZE)
 		return -EINVAL;
@@ -6538,10 +6715,29 @@ raid5_store_stripe_size(struct mddev  *mddev, const char *page, size_t len)
 	pr_debug("md/raid: change stripe_size from %lu to %lu\n",
 			conf->stripe_size, new);
 
+	if (mddev->sync_thread ||
+		test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
+		mddev->reshape_position != MaxSector ||
+		mddev->sysfs_active) {
+		err = -EBUSY;
+		goto out_unlock;
+	}
+
 	mddev_suspend(mddev);
+	mutex_lock(&conf->cache_size_mutex);
+	size = conf->max_nr_stripes;
+
+	shrink_stripes(conf);
+
 	conf->stripe_size = new;
 	conf->stripe_shift = ilog2(new) - 9;
 	conf->stripe_sectors = new >> 9;
+	if (grow_stripes(conf, size)) {
+		pr_warn("md/raid:%s: couldn't allocate buffers\n",
+				mdname(mddev));
+		err = -ENOMEM;
+	}
+	mutex_unlock(&conf->cache_size_mutex);
 	mddev_resume(mddev);
 
 out_unlock:
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 16fc29472f5c..5c05acf20e1f 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -195,6 +195,7 @@ enum reconstruct_states {
 	reconstruct_state_result,
 };
 
+#define DEFAULT_STRIPE_SIZE	4096
 struct stripe_head {
 	struct hlist_node	hash;
 	struct list_head	lru;	      /* inactive_list or handle_list */
@@ -246,6 +247,13 @@ struct stripe_head {
 		int 		     target, target2;
 		enum sum_check_flags zero_sum_result;
 	} ops;
+
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+	/* These pages will be used by bios in dev[i] */
+	struct page	**pages;
+	int	nr_pages;	/* page array size */
+	int	stripes_per_page;
+#endif
 	struct r5dev {
 		/* rreq and rvec are used for the replacement device when
 		 * writing data to both devices.
@@ -253,6 +261,7 @@ struct stripe_head {
 		struct bio	req, rreq;
 		struct bio_vec	vec, rvec;
 		struct page	*page, *orig_page;
+		unsigned int    offset;     /* offset of the page */
 		struct bio	*toread, *read, *towrite, *written;
 		sector_t	sector;			/* sector of this page */
 		unsigned long	flags;
@@ -472,7 +481,6 @@ struct disk_info {
  */
 
 #define NR_STRIPES		256
-#define DEFAULT_STRIPE_SIZE	4096
 
 #if PAGE_SIZE == DEFAULT_STRIPE_SIZE
 #define STRIPE_SIZE		PAGE_SIZE
@@ -771,6 +779,25 @@ static inline int algorithm_is_DDF(int layout)
 	return layout >= 8 && layout <= 10;
 }
 
+#if PAGE_SIZE != DEFAULT_STRIPE_SIZE
+/*
+ * Return offset of the corresponding page for r5dev.
+ */
+static inline int raid5_get_page_offset(struct stripe_head *sh, int disk_idx)
+{
+	return (disk_idx % sh->stripes_per_page) * RAID5_STRIPE_SIZE(sh->raid_conf);
+}
+
+/*
+ * Return corresponding page address for r5dev.
+ */
+static inline struct page *
+raid5_get_dev_page(struct stripe_head *sh, int disk_idx)
+{
+	return sh->pages[disk_idx / sh->stripes_per_page];
+}
+#endif
+
 extern void md_raid5_kick_device(struct r5conf *conf);
 extern int raid5_set_cache_size(struct mddev *mddev, int size);
 extern sector_t raid5_compute_blocknr(struct stripe_head *sh, int i, int previous);
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index d2397cfe178f..56e2a22e8a02 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -89,7 +89,6 @@ static dev_t nvme_chr_devt;
 static struct class *nvme_class;
 static struct class *nvme_subsys_class;
 
-static int _nvme_revalidate_disk(struct gendisk *disk);
 static void nvme_put_subsystem(struct nvme_subsystem *subsys);
 static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl,
 					   unsigned nsid);
@@ -968,10 +967,10 @@ static u32 nvme_known_admin_effects(u8 opcode)
 {
 	switch (opcode) {
 	case nvme_admin_format_nvm:
-		return NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC |
+		return NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_NCC |
 			NVME_CMD_EFFECTS_CSE_MASK;
 	case nvme_admin_sanitize_nvm:
-		return NVME_CMD_EFFECTS_CSE_MASK;
+		return NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_CSE_MASK;
 	default:
 		break;
 	}
@@ -1009,7 +1008,7 @@ static u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	 * For simplicity, IO to all namespaces is quiesced even if the command
 	 * effects say only one namespace is affected.
 	 */
-	if (effects & (NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_CSE_MASK)) {
+	if (effects & NVME_CMD_EFFECTS_CSE_MASK) {
 		mutex_lock(&ctrl->scan_lock);
 		mutex_lock(&ctrl->subsys->lock);
 		nvme_mpath_start_freeze(ctrl->subsys);
@@ -1020,36 +1019,9 @@ static u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	return effects;
 }
 
-static void nvme_update_formats(struct nvme_ctrl *ctrl, u32 *effects)
-{
-	struct nvme_ns *ns;
-
-	down_read(&ctrl->namespaces_rwsem);
-	list_for_each_entry(ns, &ctrl->namespaces, list)
-		if (_nvme_revalidate_disk(ns->disk))
-			nvme_set_queue_dying(ns);
-		else if (blk_queue_is_zoned(ns->disk->queue)) {
-			/*
-			 * IO commands are required to fully revalidate a zoned
-			 * device. Force the command effects to trigger rescan
-			 * work so report zones can run in a context with
-			 * unfrozen IO queues.
-			 */
-			*effects |= NVME_CMD_EFFECTS_NCC;
-		}
-	up_read(&ctrl->namespaces_rwsem);
-}
-
 static void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects)
 {
-	/*
-	 * Revalidate LBA changes prior to unfreezing. This is necessary to
-	 * prevent memory corruption if a logical block size was changed by
-	 * this command.
-	 */
-	if (effects & NVME_CMD_EFFECTS_LBCC)
-		nvme_update_formats(ctrl, &effects);
-	if (effects & (NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_CSE_MASK)) {
+	if (effects & NVME_CMD_EFFECTS_CSE_MASK) {
 		nvme_unfreeze(ctrl);
 		nvme_mpath_unfreeze(ctrl->subsys);
 		mutex_unlock(&ctrl->subsys->lock);
@@ -1309,6 +1281,8 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid,
 	int status, pos, len;
 	void *data;
 
+	if (ctrl->vs < NVME_VS(1, 3, 0) && !nvme_multi_css(ctrl))
+		return 0;
 	if (ctrl->quirks & NVME_QUIRK_NO_NS_DESC_LIST)
 		return 0;
 
@@ -1352,19 +1326,8 @@ free_data:
 	return status;
 }
 
-static int nvme_identify_ns_list(struct nvme_ctrl *dev, unsigned nsid, __le32 *ns_list)
-{
-	struct nvme_command c = { };
-
-	c.identify.opcode = nvme_admin_identify;
-	c.identify.cns = NVME_ID_CNS_NS_ACTIVE_LIST;
-	c.identify.nsid = cpu_to_le32(nsid);
-	return nvme_submit_sync_cmd(dev->admin_q, &c, ns_list,
-				    NVME_IDENTIFY_DATA_SIZE);
-}
-
-static int nvme_identify_ns(struct nvme_ctrl *ctrl,
-		unsigned nsid, struct nvme_id_ns **id)
+static int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid,
+			struct nvme_ns_ids *ids, struct nvme_id_ns **id)
 {
 	struct nvme_command c = { };
 	int error;
@@ -1381,9 +1344,24 @@ static int nvme_identify_ns(struct nvme_ctrl *ctrl,
 	error = nvme_submit_sync_cmd(ctrl->admin_q, &c, *id, sizeof(**id));
 	if (error) {
 		dev_warn(ctrl->device, "Identify namespace failed (%d)\n", error);
-		kfree(*id);
+		goto out_free_id;
 	}
 
+	error = -ENODEV;
+	if ((*id)->ncap == 0) /* namespace not allocated or attached */
+		goto out_free_id;
+
+	if (ctrl->vs >= NVME_VS(1, 1, 0) &&
+	    !memchr_inv(ids->eui64, 0, sizeof(ids->eui64)))
+		memcpy(ids->eui64, (*id)->eui64, sizeof(ids->eui64));
+	if (ctrl->vs >= NVME_VS(1, 2, 0) &&
+	    !memchr_inv(ids->nguid, 0, sizeof(ids->nguid)))
+		memcpy(ids->nguid, (*id)->nguid, sizeof(ids->nguid));
+
+	return 0;
+
+out_free_id:
+	kfree(*id);
 	return error;
 }
 
@@ -1905,20 +1883,6 @@ static void nvme_config_write_zeroes(struct gendisk *disk, struct nvme_ns *ns)
 					   nvme_lba_to_sect(ns, max_blocks));
 }
 
-static int nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid,
-		struct nvme_id_ns *id, struct nvme_ns_ids *ids)
-{
-	memset(ids, 0, sizeof(*ids));
-
-	if (ctrl->vs >= NVME_VS(1, 1, 0))
-		memcpy(ids->eui64, id->eui64, sizeof(id->eui64));
-	if (ctrl->vs >= NVME_VS(1, 2, 0))
-		memcpy(ids->nguid, id->nguid, sizeof(id->nguid));
-	if (ctrl->vs >= NVME_VS(1, 3, 0) || nvme_multi_css(ctrl))
-		return nvme_identify_ns_descs(ctrl, nsid, ids);
-	return 0;
-}
-
 static bool nvme_ns_ids_valid(struct nvme_ns_ids *ids)
 {
 	return !uuid_is_null(&ids->uuid) ||
@@ -1959,6 +1923,68 @@ static int nvme_setup_streams_ns(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	return 0;
 }
 
+static int nvme_configure_metadata(struct nvme_ns *ns, struct nvme_id_ns *id)
+{
+	struct nvme_ctrl *ctrl = ns->ctrl;
+
+	/*
+	 * The PI implementation requires the metadata size to be equal to the
+	 * t10 pi tuple size.
+	 */
+	ns->ms = le16_to_cpu(id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ms);
+	if (ns->ms == sizeof(struct t10_pi_tuple))
+		ns->pi_type = id->dps & NVME_NS_DPS_PI_MASK;
+	else
+		ns->pi_type = 0;
+
+	ns->features &= ~(NVME_NS_METADATA_SUPPORTED | NVME_NS_EXT_LBAS);
+	if (!ns->ms || !(ctrl->ops->flags & NVME_F_METADATA_SUPPORTED))
+		return 0;
+	if (ctrl->ops->flags & NVME_F_FABRICS) {
+		/*
+		 * The NVMe over Fabrics specification only supports metadata as
+		 * part of the extended data LBA.  We rely on HCA/HBA support to
+		 * remap the separate metadata buffer from the block layer.
+		 */
+		if (WARN_ON_ONCE(!(id->flbas & NVME_NS_FLBAS_META_EXT)))
+			return -EINVAL;
+		if (ctrl->max_integrity_segments)
+			ns->features |=
+				(NVME_NS_METADATA_SUPPORTED | NVME_NS_EXT_LBAS);
+	} else {
+		/*
+		 * For PCIe controllers, we can't easily remap the separate
+		 * metadata buffer from the block layer and thus require a
+		 * separate metadata buffer for block layer metadata/PI support.
+		 * We allow extended LBAs for the passthrough interface, though.
+		 */
+		if (id->flbas & NVME_NS_FLBAS_META_EXT)
+			ns->features |= NVME_NS_EXT_LBAS;
+		else
+			ns->features |= NVME_NS_METADATA_SUPPORTED;
+	}
+
+	return 0;
+}
+
+static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
+		struct request_queue *q)
+{
+	bool vwc = ctrl->vwc & NVME_CTRL_VWC_PRESENT;
+
+	if (ctrl->max_hw_sectors) {
+		u32 max_segments =
+			(ctrl->max_hw_sectors / (NVME_CTRL_PAGE_SIZE >> 9)) + 1;
+
+		max_segments = min_not_zero(max_segments, ctrl->max_segments);
+		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
+		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
+	}
+	blk_queue_virt_boundary(q, NVME_CTRL_PAGE_SIZE - 1);
+	blk_queue_dma_alignment(q, 7);
+	blk_queue_write_cache(q, vwc, vwc);
+}
+
 static void nvme_update_disk_info(struct gendisk *disk,
 		struct nvme_ns *ns, struct nvme_id_ns *id)
 {
@@ -1966,11 +1992,15 @@ static void nvme_update_disk_info(struct gendisk *disk,
 	unsigned short bs = 1 << ns->lba_shift;
 	u32 atomic_bs, phys_bs, io_opt = 0;
 
+	/*
+	 * The block layer can't support LBA sizes larger than the page size
+	 * yet, so catch this early and don't allow block I/O.
+	 */
 	if (ns->lba_shift > PAGE_SHIFT) {
-		/* unsupported block size, set capacity to 0 later */
+		capacity = 0;
 		bs = (1 << 9);
 	}
-	blk_mq_freeze_queue(disk->queue);
+
 	blk_integrity_unregister(disk);
 
 	atomic_bs = phys_bs = bs;
@@ -2005,13 +2035,6 @@ static void nvme_update_disk_info(struct gendisk *disk,
 	blk_queue_io_opt(disk->queue, io_opt);
 
 	/*
-	 * The block layer can't support LBA sizes larger than the page size
-	 * yet, so catch this early and don't allow block I/O.
-	 */
-	if (ns->lba_shift > PAGE_SHIFT)
-		capacity = 0;
-
-	/*
 	 * Register a metadata profile for PI, or the plain non-integrity NVMe
 	 * metadata masquerading as Type 0 if supported, otherwise reject block
 	 * I/O to namespaces with metadata except when the namespace supports
@@ -2035,8 +2058,6 @@ static void nvme_update_disk_info(struct gendisk *disk,
 		set_disk_ro(disk, true);
 	else
 		set_disk_ro(disk, false);
-
-	blk_mq_unfreeze_queue(disk->queue);
 }
 
 static inline bool nvme_first_scan(struct gendisk *disk)
@@ -2076,151 +2097,49 @@ static void nvme_set_chunk_sectors(struct nvme_ns *ns, struct nvme_id_ns *id)
 	blk_queue_chunk_sectors(ns->queue, iob);
 }
 
-static int __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
+static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_id_ns *id)
 {
 	unsigned lbaf = id->flbas & NVME_NS_FLBAS_LBA_MASK;
-	struct nvme_ns *ns = disk->private_data;
-	struct nvme_ctrl *ctrl = ns->ctrl;
 	int ret;
 
-	/*
-	 * If identify namespace failed, use default 512 byte block size so
-	 * block layer can use before failing read/write for 0 capacity.
-	 */
+	blk_mq_freeze_queue(ns->disk->queue);
 	ns->lba_shift = id->lbaf[lbaf].ds;
-	if (ns->lba_shift == 0)
-		ns->lba_shift = 9;
+	nvme_set_queue_limits(ns->ctrl, ns->queue);
 
-	switch (ns->head->ids.csi) {
-	case NVME_CSI_NVM:
-		break;
-	case NVME_CSI_ZNS:
-		ret = nvme_update_zone_info(disk, ns, lbaf);
-		if (ret) {
-			dev_warn(ctrl->device,
-				"failed to add zoned namespace:%u ret:%d\n",
-				ns->head->ns_id, ret);
-			return ret;
-		}
-		break;
-	default:
-		dev_warn(ctrl->device, "unknown csi:%u ns:%u\n",
-			ns->head->ids.csi, ns->head->ns_id);
-		return -ENODEV;
+	if (ns->head->ids.csi == NVME_CSI_ZNS) {
+		ret = nvme_update_zone_info(ns, lbaf);
+		if (ret)
+			goto out_unfreeze;
 	}
 
-	ns->features = 0;
-	ns->ms = le16_to_cpu(id->lbaf[lbaf].ms);
-	/* the PI implementation requires metadata equal t10 pi tuple size */
-	if (ns->ms == sizeof(struct t10_pi_tuple))
-		ns->pi_type = id->dps & NVME_NS_DPS_PI_MASK;
-	else
-		ns->pi_type = 0;
+	ret = nvme_configure_metadata(ns, id);
+	if (ret)
+		goto out_unfreeze;
+	nvme_set_chunk_sectors(ns, id);
+	nvme_update_disk_info(ns->disk, ns, id);
+	blk_mq_unfreeze_queue(ns->disk->queue);
 
-	if (ns->ms) {
-		/*
-		 * For PCIe only the separate metadata pointer is supported,
-		 * as the block layer supplies metadata in a separate bio_vec
-		 * chain. For Fabrics, only metadata as part of extended data
-		 * LBA is supported on the wire per the Fabrics specification,
-		 * but the HBA/HCA will do the remapping from the separate
-		 * metadata buffers for us.
-		 */
-		if (id->flbas & NVME_NS_FLBAS_META_EXT) {
-			ns->features |= NVME_NS_EXT_LBAS;
-			if ((ctrl->ops->flags & NVME_F_FABRICS) &&
-			    (ctrl->ops->flags & NVME_F_METADATA_SUPPORTED) &&
-			    ctrl->max_integrity_segments)
-				ns->features |= NVME_NS_METADATA_SUPPORTED;
-		} else {
-			if (WARN_ON_ONCE(ctrl->ops->flags & NVME_F_FABRICS))
-				return -EINVAL;
-			if (ctrl->ops->flags & NVME_F_METADATA_SUPPORTED)
-				ns->features |= NVME_NS_METADATA_SUPPORTED;
-		}
+	if (blk_queue_is_zoned(ns->queue)) {
+		ret = nvme_revalidate_zones(ns);
+		if (ret)
+			return ret;
 	}
 
-	nvme_set_chunk_sectors(ns, id);
-	nvme_update_disk_info(disk, ns, id);
 #ifdef CONFIG_NVME_MULTIPATH
 	if (ns->head->disk) {
+		blk_mq_freeze_queue(ns->head->disk->queue);
 		nvme_update_disk_info(ns->head->disk, ns, id);
 		blk_stack_limits(&ns->head->disk->queue->limits,
 				 &ns->queue->limits, 0);
 		blk_queue_update_readahead(ns->head->disk->queue);
 		nvme_update_bdev_size(ns->head->disk);
+		blk_mq_unfreeze_queue(ns->head->disk->queue);
 	}
 #endif
 	return 0;
-}
-
-static int _nvme_revalidate_disk(struct gendisk *disk)
-{
-	struct nvme_ns *ns = disk->private_data;
-	struct nvme_ctrl *ctrl = ns->ctrl;
-	struct nvme_id_ns *id;
-	struct nvme_ns_ids ids;
-	int ret = 0;
-
-	if (test_bit(NVME_NS_DEAD, &ns->flags)) {
-		set_capacity(disk, 0);
-		return -ENODEV;
-	}
 
-	ret = nvme_identify_ns(ctrl, ns->head->ns_id, &id);
-	if (ret)
-		goto out;
-
-	if (id->ncap == 0) {
-		ret = -ENODEV;
-		goto free_id;
-	}
-
-	ret = nvme_report_ns_ids(ctrl, ns->head->ns_id, id, &ids);
-	if (ret)
-		goto free_id;
-
-	if (!nvme_ns_ids_equal(&ns->head->ids, &ids)) {
-		dev_err(ctrl->device,
-			"identifiers changed for nsid %d\n", ns->head->ns_id);
-		ret = -ENODEV;
-		goto free_id;
-	}
-
-	ret = __nvme_revalidate_disk(disk, id);
-free_id:
-	kfree(id);
-out:
-	/*
-	 * Only fail the function if we got a fatal error back from the
-	 * device, otherwise ignore the error and just move on.
-	 */
-	if (ret == -ENOMEM || (ret > 0 && !(ret & NVME_SC_DNR)))
-		ret = 0;
-	else if (ret > 0)
-		ret = blk_status_to_errno(nvme_error_status(ret));
-	return ret;
-}
-
-static int nvme_revalidate_disk(struct gendisk *disk)
-{
-	int ret;
-
-	ret = _nvme_revalidate_disk(disk);
-	if (ret)
-		return ret;
-
-#ifdef CONFIG_BLK_DEV_ZONED
-	if (blk_queue_is_zoned(disk->queue)) {
-		struct nvme_ns *ns = disk->private_data;
-		struct nvme_ctrl *ctrl = ns->ctrl;
-
-		ret = blk_revalidate_disk_zones(disk, NULL);
-		if (!ret)
-			blk_queue_max_zone_append_sectors(disk->queue,
-							  ctrl->max_zone_append);
-	}
-#endif
+out_unfreeze:
+	blk_mq_unfreeze_queue(ns->disk->queue);
 	return ret;
 }
 
@@ -2502,26 +2421,6 @@ int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl)
 }
 EXPORT_SYMBOL_GPL(nvme_shutdown_ctrl);
 
-static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
-		struct request_queue *q)
-{
-	bool vwc = false;
-
-	if (ctrl->max_hw_sectors) {
-		u32 max_segments =
-			(ctrl->max_hw_sectors / (NVME_CTRL_PAGE_SIZE >> 9)) + 1;
-
-		max_segments = min_not_zero(max_segments, ctrl->max_segments);
-		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
-		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
-	}
-	blk_queue_virt_boundary(q, NVME_CTRL_PAGE_SIZE - 1);
-	blk_queue_dma_alignment(q, 7);
-	if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
-		vwc = true;
-	blk_queue_write_cache(q, vwc, vwc);
-}
-
 static int nvme_configure_timestamp(struct nvme_ctrl *ctrl)
 {
 	__le64 ts;
@@ -3025,26 +2924,10 @@ int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, u8 csi,
 	return nvme_submit_sync_cmd(ctrl->admin_q, &c, log, size);
 }
 
-static struct nvme_cel *nvme_find_cel(struct nvme_ctrl *ctrl, u8 csi)
-{
-	struct nvme_cel *cel, *ret = NULL;
-
-	spin_lock_irq(&ctrl->lock);
-	list_for_each_entry(cel, &ctrl->cels, entry) {
-		if (cel->csi == csi) {
-			ret = cel;
-			break;
-		}
-	}
-	spin_unlock_irq(&ctrl->lock);
-
-	return ret;
-}
-
 static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi,
 				struct nvme_effects_log **log)
 {
-	struct nvme_cel *cel = nvme_find_cel(ctrl, csi);
+	struct nvme_cel *cel = xa_load(&ctrl->cels, csi);
 	int ret;
 
 	if (cel)
@@ -3062,10 +2945,7 @@ static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi,
 	}
 
 	cel->csi = csi;
-
-	spin_lock_irq(&ctrl->lock);
-	list_add_tail(&cel->entry, &ctrl->cels);
-	spin_unlock_irq(&ctrl->lock);
+	xa_store(&ctrl->cels, cel->csi, cel, GFP_KERNEL);
 out:
 	*log = &cel->log;
 	return 0;
@@ -3846,25 +3726,16 @@ out:
 }
 
 static int nvme_init_ns_head(struct nvme_ns *ns, unsigned nsid,
-		struct nvme_id_ns *id)
+		struct nvme_ns_ids *ids, bool is_shared)
 {
 	struct nvme_ctrl *ctrl = ns->ctrl;
-	bool is_shared = id->nmic & NVME_NS_NMIC_SHARED;
 	struct nvme_ns_head *head = NULL;
-	struct nvme_ns_ids ids;
 	int ret = 0;
 
-	ret = nvme_report_ns_ids(ctrl, nsid, id, &ids);
-	if (ret) {
-		if (ret < 0)
-			return ret;
-		return blk_status_to_errno(nvme_error_status(ret));
-	}
-
 	mutex_lock(&ctrl->subsys->lock);
 	head = nvme_find_ns_head(ctrl->subsys, nsid);
 	if (!head) {
-		head = nvme_alloc_ns_head(ctrl, nsid, &ids);
+		head = nvme_alloc_ns_head(ctrl, nsid, ids);
 		if (IS_ERR(head)) {
 			ret = PTR_ERR(head);
 			goto out_unlock;
@@ -3877,7 +3748,7 @@ static int nvme_init_ns_head(struct nvme_ns *ns, unsigned nsid,
 				"Duplicate unshared namespace %d\n", nsid);
 			goto out_put_ns_head;
 		}
-		if (!nvme_ns_ids_equal(&head->ids, &ids)) {
+		if (!nvme_ns_ids_equal(&head->ids, ids)) {
 			dev_err(ctrl->device,
 				"IDs don't match for shared namespace %d\n",
 					nsid);
@@ -3925,7 +3796,8 @@ struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 }
 EXPORT_SYMBOL_NS_GPL(nvme_find_get_ns, NVME_TARGET_PASSTHRU);
 
-static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
+static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid,
+		struct nvme_ns_ids *ids)
 {
 	struct nvme_ns *ns;
 	struct gendisk *disk;
@@ -3933,9 +3805,12 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	char disk_name[DISK_NAME_LEN];
 	int node = ctrl->numa_node, flags = GENHD_FL_EXT_DEVT, ret;
 
+	if (nvme_identify_ns(ctrl, nsid, ids, &id))
+		return;
+
 	ns = kzalloc_node(sizeof(*ns), GFP_KERNEL, node);
 	if (!ns)
-		return;
+		goto out_free_id;
 
 	ns->queue = blk_mq_init_queue(ctrl->tagset);
 	if (IS_ERR(ns->queue))
@@ -3950,23 +3825,11 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 
 	ns->queue->queuedata = ns;
 	ns->ctrl = ctrl;
-
 	kref_init(&ns->kref);
-	ns->lba_shift = 9; /* set to a default value for 512 until disk is validated */
-
-	blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift);
-	nvme_set_queue_limits(ctrl, ns->queue);
 
-	ret = nvme_identify_ns(ctrl, nsid, &id);
+	ret = nvme_init_ns_head(ns, nsid, ids, id->nmic & NVME_NS_NMIC_SHARED);
 	if (ret)
 		goto out_free_queue;
-
-	if (id->ncap == 0)	/* no namespace (legacy quirk) */
-		goto out_free_id;
-
-	ret = nvme_init_ns_head(ns, nsid, id);
-	if (ret)
-		goto out_free_id;
 	nvme_set_disk_name(disk_name, ns, ctrl, &flags);
 
 	disk = alloc_disk_node(0, node);
@@ -3980,7 +3843,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	memcpy(disk->disk_name, disk_name, DISK_NAME_LEN);
 	ns->disk = disk;
 
-	if (__nvme_revalidate_disk(disk, id))
+	if (nvme_update_ns_info(ns, id))
 		goto out_put_disk;
 
 	if ((ctrl->quirks & NVME_QUIRK_LIGHTNVM) && id->vs[0] == 0x1) {
@@ -4015,12 +3878,12 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 		list_del_init(&ns->head->entry);
 	mutex_unlock(&ctrl->subsys->lock);
 	nvme_put_ns_head(ns->head);
- out_free_id:
-	kfree(id);
  out_free_queue:
 	blk_cleanup_queue(ns->queue);
  out_free_ns:
 	kfree(ns);
+ out_free_id:
+	kfree(id);
 }
 
 static void nvme_ns_remove(struct nvme_ns *ns)
@@ -4028,6 +3891,7 @@ static void nvme_ns_remove(struct nvme_ns *ns)
 	if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
 		return;
 
+	set_capacity(ns->disk, 0);
 	nvme_fault_inject_fini(&ns->fault_inject);
 
 	mutex_lock(&ns->ctrl->subsys->lock);
@@ -4065,22 +3929,75 @@ static void nvme_ns_remove_by_nsid(struct nvme_ctrl *ctrl, u32 nsid)
 	}
 }
 
-static void nvme_validate_ns(struct nvme_ctrl *ctrl, unsigned nsid)
+static void nvme_validate_ns(struct nvme_ns *ns, struct nvme_ns_ids *ids)
+{
+	struct nvme_id_ns *id;
+	int ret = -ENODEV;
+
+	if (test_bit(NVME_NS_DEAD, &ns->flags))
+		goto out;
+
+	ret = nvme_identify_ns(ns->ctrl, ns->head->ns_id, ids, &id);
+	if (ret)
+		goto out;
+
+	ret = -ENODEV;
+	if (!nvme_ns_ids_equal(&ns->head->ids, ids)) {
+		dev_err(ns->ctrl->device,
+			"identifiers changed for nsid %d\n", ns->head->ns_id);
+		goto out_free_id;
+	}
+
+	ret = nvme_update_ns_info(ns, id);
+
+out_free_id:
+	kfree(id);
+out:
+	/*
+	 * Only remove the namespace if we got a fatal error back from the
+	 * device, otherwise ignore the error and just move on.
+	 *
+	 * TODO: we should probably schedule a delayed retry here.
+	 */
+	if (ret && ret != -ENOMEM && !(ret > 0 && !(ret & NVME_SC_DNR)))
+		nvme_ns_remove(ns);
+	else
+		revalidate_disk_size(ns->disk, true);
+}
+
+static void nvme_validate_or_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 {
+	struct nvme_ns_ids ids = { };
 	struct nvme_ns *ns;
-	int ret;
+
+	if (nvme_identify_ns_descs(ctrl, nsid, &ids))
+		return;
 
 	ns = nvme_find_get_ns(ctrl, nsid);
-	if (!ns) {
-		nvme_alloc_ns(ctrl, nsid);
+	if (ns) {
+		nvme_validate_ns(ns, &ids);
+		nvme_put_ns(ns);
 		return;
 	}
 
-	ret = nvme_revalidate_disk(ns->disk);
-	revalidate_disk_size(ns->disk, ret == 0);
-	if (ret)
-		nvme_ns_remove(ns);
-	nvme_put_ns(ns);
+	switch (ids.csi) {
+	case NVME_CSI_NVM:
+		nvme_alloc_ns(ctrl, nsid, &ids);
+		break;
+	case NVME_CSI_ZNS:
+		if (!IS_ENABLED(CONFIG_BLK_DEV_ZONED)) {
+			dev_warn(ctrl->device,
+				"nsid %u not supported without CONFIG_BLK_DEV_ZONED\n",
+				nsid);
+			break;
+		}
+		nvme_alloc_ns(ctrl, nsid, &ids);
+		break;
+	default:
+		dev_warn(ctrl->device, "unknown csi %u for nsid %u\n",
+			ids.csi, nsid);
+		break;
+	}
 }
 
 static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl,
@@ -4116,7 +4033,14 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl)
 		return -ENOMEM;
 
 	for (;;) {
-		ret = nvme_identify_ns_list(ctrl, prev, ns_list);
+		struct nvme_command cmd = {
+			.identify.opcode	= nvme_admin_identify,
+			.identify.cns		= NVME_ID_CNS_NS_ACTIVE_LIST,
+			.identify.nsid		= cpu_to_le32(prev),
+		};
+
+		ret = nvme_submit_sync_cmd(ctrl->admin_q, &cmd, ns_list,
+					    NVME_IDENTIFY_DATA_SIZE);
 		if (ret)
 			goto free;
 
@@ -4125,7 +4049,7 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl)
 
 			if (!nsid)	/* end of the list? */
 				goto out;
-			nvme_validate_ns(ctrl, nsid);
+			nvme_validate_or_alloc_ns(ctrl, nsid);
 			while (++prev < nsid)
 				nvme_ns_remove_by_nsid(ctrl, prev);
 		}
@@ -4148,7 +4072,7 @@ static void nvme_scan_ns_sequential(struct nvme_ctrl *ctrl)
 	kfree(id);
 
 	for (i = 1; i <= nn; i++)
-		nvme_validate_ns(ctrl, i);
+		nvme_validate_or_alloc_ns(ctrl, i);
 
 	nvme_remove_invalid_namespaces(ctrl, nn);
 }
@@ -4453,15 +4377,11 @@ static void nvme_free_ctrl(struct device *dev)
 	struct nvme_ctrl *ctrl =
 		container_of(dev, struct nvme_ctrl, ctrl_device);
 	struct nvme_subsystem *subsys = ctrl->subsys;
-	struct nvme_cel *cel, *next;
 
 	if (!subsys || ctrl->instance != subsys->instance)
 		ida_simple_remove(&nvme_instance_ida, ctrl->instance);
 
-	list_for_each_entry_safe(cel, next, &ctrl->cels, entry) {
-		list_del(&cel->entry);
-		kfree(cel);
-	}
+	xa_destroy(&ctrl->cels);
 
 	nvme_mpath_uninit(ctrl);
 	__free_page(ctrl->discard_page);
@@ -4493,7 +4413,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 	spin_lock_init(&ctrl->lock);
 	mutex_init(&ctrl->scan_lock);
 	INIT_LIST_HEAD(&ctrl->namespaces);
-	INIT_LIST_HEAD(&ctrl->cels);
+	xa_init(&ctrl->cels);
 	init_rwsem(&ctrl->namespaces_rwsem);
 	ctrl->dev = dev;
 	ctrl->ops = ops;
@@ -4673,28 +4593,13 @@ void nvme_sync_queues(struct nvme_ctrl *ctrl)
 }
 EXPORT_SYMBOL_GPL(nvme_sync_queues);
 
-struct nvme_ctrl *nvme_ctrl_get_by_path(const char *path)
+struct nvme_ctrl *nvme_ctrl_from_file(struct file *file)
 {
-	struct nvme_ctrl *ctrl;
-	struct file *f;
-
-	f = filp_open(path, O_RDWR, 0);
-	if (IS_ERR(f))
-		return ERR_CAST(f);
-
-	if (f->f_op != &nvme_dev_fops) {
-		ctrl = ERR_PTR(-EINVAL);
-		goto out_close;
-	}
-
-	ctrl = f->private_data;
-	nvme_get_ctrl(ctrl);
-
-out_close:
-	filp_close(f, NULL);
-	return ctrl;
+	if (file->f_op != &nvme_dev_fops)
+		return NULL;
+	return file->private_data;
 }
-EXPORT_SYMBOL_NS_GPL(nvme_ctrl_get_by_path, NVME_TARGET_PASSTHRU);
+EXPORT_SYMBOL_NS_GPL(nvme_ctrl_from_file, NVME_TARGET_PASSTHRU);
 
 /*
  * Check we didn't inadvertently grow the command structure sizes:
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 8ac37430347c..e7c88b40f5bb 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -300,7 +300,7 @@ struct nvme_ctrl {
 	unsigned long quirks;
 	struct nvme_id_power_state psd[32];
 	struct nvme_effects_log *effects;
-	struct list_head cels;
+	struct xarray cels;
 	struct work_struct scan_work;
 	struct work_struct async_event_work;
 	struct delayed_work ka_work;
@@ -758,10 +758,9 @@ static inline void nvme_mpath_start_freeze(struct nvme_subsystem *subsys)
 }
 #endif /* CONFIG_NVME_MULTIPATH */
 
+int nvme_revalidate_zones(struct nvme_ns *ns);
 #ifdef CONFIG_BLK_DEV_ZONED
-int nvme_update_zone_info(struct gendisk *disk, struct nvme_ns *ns,
-			  unsigned lbaf);
-
+int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf);
 int nvme_report_zones(struct gendisk *disk, sector_t sector,
 		      unsigned int nr_zones, report_zones_cb cb, void *data);
 
@@ -778,9 +777,7 @@ static inline blk_status_t nvme_setup_zone_mgmt_send(struct nvme_ns *ns,
 	return BLK_STS_NOTSUPP;
 }
 
-static inline int nvme_update_zone_info(struct gendisk *disk,
-					struct nvme_ns *ns,
-					unsigned lbaf)
+static inline int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf)
 {
 	dev_warn(ns->ctrl->device,
 		 "Please enable CONFIG_BLK_DEV_ZONED to support ZNS devices\n");
@@ -825,7 +822,7 @@ static inline int nvme_hwmon_init(struct nvme_ctrl *ctrl)
 u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 			 u8 opcode);
 void nvme_execute_passthru_rq(struct request *rq);
-struct nvme_ctrl *nvme_ctrl_get_by_path(const char *path);
+struct nvme_ctrl *nvme_ctrl_from_file(struct file *file);
 struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid);
 void nvme_put_ns(struct nvme_ns *ns);
 
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 8984796db0c8..e5b02242f3ca 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -2038,32 +2038,30 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues)
 		.calc_sets	= nvme_calc_irq_sets,
 		.priv		= dev,
 	};
-	unsigned int irq_queues, this_p_queues;
+	unsigned int irq_queues, poll_queues;
 
 	/*
-	 * Poll queues don't need interrupts, but we need at least one IO
-	 * queue left over for non-polled IO.
+	 * Poll queues don't need interrupts, but we need at least one I/O queue
+	 * left over for non-polled I/O.
 	 */
-	this_p_queues = dev->nr_poll_queues;
-	if (this_p_queues >= nr_io_queues) {
-		this_p_queues = nr_io_queues - 1;
-		irq_queues = 1;
-	} else {
-		irq_queues = nr_io_queues - this_p_queues + 1;
-	}
-	dev->io_queues[HCTX_TYPE_POLL] = this_p_queues;
+	poll_queues = min(dev->nr_poll_queues, nr_io_queues - 1);
+	dev->io_queues[HCTX_TYPE_POLL] = poll_queues;
 
-	/* Initialize for the single interrupt case */
+	/*
+	 * Initialize for the single interrupt case, will be updated in
+	 * nvme_calc_irq_sets().
+	 */
 	dev->io_queues[HCTX_TYPE_DEFAULT] = 1;
 	dev->io_queues[HCTX_TYPE_READ] = 0;
 
 	/*
-	 * Some Apple controllers require all queues to use the
-	 * first vector.
+	 * We need interrupts for the admin queue and each non-polled I/O queue,
+	 * but some Apple controllers require all queues to use the first
+	 * vector.
 	 */
-	if (dev->ctrl.quirks & NVME_QUIRK_SINGLE_VECTOR)
-		irq_queues = 1;
-
+	irq_queues = 1;
+	if (!(dev->ctrl.quirks & NVME_QUIRK_SINGLE_VECTOR))
+		irq_queues += (nr_io_queues - poll_queues);
 	return pci_alloc_irq_vectors_affinity(pdev, 1, irq_queues,
 			      PCI_IRQ_ALL_TYPES | PCI_IRQ_AFFINITY, &affd);
 }
@@ -3187,7 +3185,6 @@ static const struct pci_device_id nvme_id_table[] = {
 				NVME_QUIRK_IGNORE_DEV_SUBNQN, },
 	{ PCI_DEVICE(0x1c5c, 0x1504),   /* SK Hynix PC400 */
 		.driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, },
-	{ PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001),
 		.driver_data = NVME_QUIRK_SINGLE_VECTOR },
 	{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2003) },
@@ -3195,6 +3192,8 @@ static const struct pci_device_id nvme_id_table[] = {
 		.driver_data = NVME_QUIRK_SINGLE_VECTOR |
 				NVME_QUIRK_128_BYTES_SQES |
 				NVME_QUIRK_SHARED_TAGS },
+
+	{ PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) },
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, nvme_id_table);
diff --git a/drivers/nvme/host/zns.c b/drivers/nvme/host/zns.c
index 57cfd78731fb..67e87e9f306f 100644
--- a/drivers/nvme/host/zns.c
+++ b/drivers/nvme/host/zns.c
@@ -7,6 +7,17 @@
 #include <linux/vmalloc.h>
 #include "nvme.h"
 
+int nvme_revalidate_zones(struct nvme_ns *ns)
+{
+	struct request_queue *q = ns->queue;
+	int ret;
+
+	ret = blk_revalidate_disk_zones(ns->disk, NULL);
+	if (!ret)
+		blk_queue_max_zone_append_sectors(q, ns->ctrl->max_zone_append);
+	return ret;
+}
+
 static int nvme_set_max_append(struct nvme_ctrl *ctrl)
 {
 	struct nvme_command c = { };
@@ -35,11 +46,10 @@ static int nvme_set_max_append(struct nvme_ctrl *ctrl)
 	return 0;
 }
 
-int nvme_update_zone_info(struct gendisk *disk, struct nvme_ns *ns,
-			  unsigned lbaf)
+int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf)
 {
 	struct nvme_effects_log *log = ns->head->effects;
-	struct request_queue *q = disk->queue;
+	struct request_queue *q = ns->queue;
 	struct nvme_command c = { };
 	struct nvme_id_ns_zns *id;
 	int status;
@@ -133,28 +143,6 @@ static void *nvme_zns_alloc_report_buffer(struct nvme_ns *ns,
 	return NULL;
 }
 
-static int __nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
-				  struct nvme_zone_report *report,
-				  size_t buflen)
-{
-	struct nvme_command c = { };
-	int ret;
-
-	c.zmr.opcode = nvme_cmd_zone_mgmt_recv;
-	c.zmr.nsid = cpu_to_le32(ns->head->ns_id);
-	c.zmr.slba = cpu_to_le64(nvme_sect_to_lba(ns, sector));
-	c.zmr.numd = cpu_to_le32(nvme_bytes_to_numd(buflen));
-	c.zmr.zra = NVME_ZRA_ZONE_REPORT;
-	c.zmr.zrasf = NVME_ZRASF_ZONE_REPORT_ALL;
-	c.zmr.pr = NVME_REPORT_ZONE_PARTIAL;
-
-	ret = nvme_submit_sync_cmd(ns->queue, &c, report, buflen);
-	if (ret)
-		return ret;
-
-	return le64_to_cpu(report->nr_zones);
-}
-
 static int nvme_zone_parse_entry(struct nvme_ns *ns,
 				 struct nvme_zone_descriptor *entry,
 				 unsigned int idx, report_zones_cb cb,
@@ -182,6 +170,7 @@ static int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
 			unsigned int nr_zones, report_zones_cb cb, void *data)
 {
 	struct nvme_zone_report *report;
+	struct nvme_command c = { };
 	int ret, zone_idx = 0;
 	unsigned int nz, i;
 	size_t buflen;
@@ -190,14 +179,26 @@ static int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
 	if (!report)
 		return -ENOMEM;
 
+	c.zmr.opcode = nvme_cmd_zone_mgmt_recv;
+	c.zmr.nsid = cpu_to_le32(ns->head->ns_id);
+	c.zmr.numd = cpu_to_le32(nvme_bytes_to_numd(buflen));
+	c.zmr.zra = NVME_ZRA_ZONE_REPORT;
+	c.zmr.zrasf = NVME_ZRASF_ZONE_REPORT_ALL;
+	c.zmr.pr = NVME_REPORT_ZONE_PARTIAL;
+
 	sector &= ~(ns->zsze - 1);
 	while (zone_idx < nr_zones && sector < get_capacity(ns->disk)) {
 		memset(report, 0, buflen);
-		ret = __nvme_ns_report_zones(ns, sector, report, buflen);
-		if (ret < 0)
+
+		c.zmr.slba = cpu_to_le64(nvme_sect_to_lba(ns, sector));
+		ret = nvme_submit_sync_cmd(ns->queue, &c, report, buflen);
+		if (ret) {
+			if (ret > 0)
+				ret = -EIO;
 			goto out_free;
+		}
 
-		nz = min_t(unsigned int, ret, nr_zones);
+		nz = min((unsigned int)le64_to_cpu(report->nr_zones), nr_zones);
 		if (!nz)
 			break;
 
diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c
index e9fe91786bbb..dca34489a1dc 100644
--- a/drivers/nvme/target/admin-cmd.c
+++ b/drivers/nvme/target/admin-cmd.c
@@ -727,7 +727,9 @@ u16 nvmet_set_feat_kato(struct nvmet_req *req)
 {
 	u32 val32 = le32_to_cpu(req->cmd->common.cdw11);
 
+	nvmet_stop_keep_alive_timer(req->sq->ctrl);
 	req->sq->ctrl->kato = DIV_ROUND_UP(val32, 1000);
+	nvmet_start_keep_alive_timer(req->sq->ctrl);
 
 	nvmet_set_result(req, req->sq->ctrl->kato);
 
diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index b7b63330b5ef..25d62d867563 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -395,7 +395,7 @@ static void nvmet_keep_alive_timer(struct work_struct *work)
 	nvmet_ctrl_fatal_error(ctrl);
 }
 
-static void nvmet_start_keep_alive_timer(struct nvmet_ctrl *ctrl)
+void nvmet_start_keep_alive_timer(struct nvmet_ctrl *ctrl)
 {
 	if (unlikely(ctrl->kato == 0))
 		return;
@@ -407,7 +407,7 @@ static void nvmet_start_keep_alive_timer(struct nvmet_ctrl *ctrl)
 	schedule_delayed_work(&ctrl->ka_work, ctrl->kato * HZ);
 }
 
-static void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl)
+void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl)
 {
 	if (unlikely(ctrl->kato == 0))
 		return;
diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c
index e6861cc10e7d..cd4e73aa9807 100644
--- a/drivers/nvme/target/fc.c
+++ b/drivers/nvme/target/fc.c
@@ -1019,7 +1019,7 @@ static void
 nvmet_fc_free_hostport(struct nvmet_fc_hostport *hostport)
 {
 	/* if LLDD not implemented, leave as NULL */
-	if (!hostport->hosthandle)
+	if (!hostport || !hostport->hosthandle)
 		return;
 
 	nvmet_fc_hostport_put(hostport);
diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c
index 0d6008cf66a2..f6d81239be21 100644
--- a/drivers/nvme/target/loop.c
+++ b/drivers/nvme/target/loop.c
@@ -579,7 +579,7 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev,
 	ret = nvme_init_ctrl(&ctrl->ctrl, dev, &nvme_loop_ctrl_ops,
 				0 /* no quirks, we're perfect! */);
 	if (ret)
-		goto out_put_ctrl;
+		goto out;
 
 	if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING))
 		WARN_ON_ONCE(1);
@@ -635,8 +635,8 @@ out_free_queues:
 	kfree(ctrl->queues);
 out_uninit_ctrl:
 	nvme_uninit_ctrl(&ctrl->ctrl);
-out_put_ctrl:
 	nvme_put_ctrl(&ctrl->ctrl);
+out:
 	if (ret > 0)
 		ret = -EIO;
 	return ERR_PTR(ret);
diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h
index 47ee3fb193bd..559a15ccc322 100644
--- a/drivers/nvme/target/nvmet.h
+++ b/drivers/nvme/target/nvmet.h
@@ -395,6 +395,8 @@ void nvmet_get_feat_async_event(struct nvmet_req *req);
 u16 nvmet_set_feat_kato(struct nvmet_req *req);
 u16 nvmet_set_feat_async_event(struct nvmet_req *req, u32 mask);
 void nvmet_execute_async_event(struct nvmet_req *req);
+void nvmet_start_keep_alive_timer(struct nvmet_ctrl *ctrl);
+void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl);
 
 u16 nvmet_parse_connect_cmd(struct nvmet_req *req);
 void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id);
diff --git a/drivers/nvme/target/passthru.c b/drivers/nvme/target/passthru.c
index dacfa7435d0b..56c571052216 100644
--- a/drivers/nvme/target/passthru.c
+++ b/drivers/nvme/target/passthru.c
@@ -456,10 +456,26 @@ u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req)
 			req->execute = nvmet_passthru_execute_cmd;
 			req->p.use_workqueue = true;
 			return NVME_SC_SUCCESS;
+		case NVME_ID_CNS_CS_CTRL:
+			switch (req->cmd->identify.csi) {
+			case NVME_CSI_ZNS:
+				req->execute = nvmet_passthru_execute_cmd;
+				req->p.use_workqueue = true;
+				return NVME_SC_SUCCESS;
+			}
+			return NVME_SC_INVALID_OPCODE | NVME_SC_DNR;
 		case NVME_ID_CNS_NS:
 			req->execute = nvmet_passthru_execute_cmd;
 			req->p.use_workqueue = true;
 			return NVME_SC_SUCCESS;
+		case NVME_ID_CNS_CS_NS:
+			switch (req->cmd->identify.csi) {
+			case NVME_CSI_ZNS:
+				req->execute = nvmet_passthru_execute_cmd;
+				req->p.use_workqueue = true;
+				return NVME_SC_SUCCESS;
+			}
+			return NVME_SC_INVALID_OPCODE | NVME_SC_DNR;
 		default:
 			return nvmet_setup_passthru_command(req);
 		}
@@ -474,6 +490,7 @@ u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req)
 int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys)
 {
 	struct nvme_ctrl *ctrl;
+	struct file *file;
 	int ret = -EINVAL;
 	void *old;
 
@@ -488,24 +505,29 @@ int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys)
 		goto out_unlock;
 	}
 
-	ctrl = nvme_ctrl_get_by_path(subsys->passthru_ctrl_path);
-	if (IS_ERR(ctrl)) {
-		ret = PTR_ERR(ctrl);
+	file = filp_open(subsys->passthru_ctrl_path, O_RDWR, 0);
+	if (IS_ERR(file)) {
+		ret = PTR_ERR(file);
+		goto out_unlock;
+	}
+
+	ctrl = nvme_ctrl_from_file(file);
+	if (!ctrl) {
 		pr_err("failed to open nvme controller %s\n",
 		       subsys->passthru_ctrl_path);
 
-		goto out_unlock;
+		goto out_put_file;
 	}
 
 	old = xa_cmpxchg(&passthru_subsystems, ctrl->cntlid, NULL,
 			 subsys, GFP_KERNEL);
 	if (xa_is_err(old)) {
 		ret = xa_err(old);
-		goto out_put_ctrl;
+		goto out_put_file;
 	}
 
 	if (old)
-		goto out_put_ctrl;
+		goto out_put_file;
 
 	subsys->passthru_ctrl = ctrl;
 	subsys->ver = ctrl->vs;
@@ -516,13 +538,12 @@ int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys)
 			NVME_TERTIARY(subsys->ver));
 		subsys->ver = NVME_VS(1, 2, 1);
 	}
-
+	nvme_get_ctrl(ctrl);
 	__module_get(subsys->passthru_ctrl->ops->module);
-	mutex_unlock(&subsys->lock);
-	return 0;
+	ret = 0;
 
-out_put_ctrl:
-	nvme_put_ctrl(ctrl);
+out_put_file:
+	filp_close(file, NULL);
 out_unlock:
 	mutex_unlock(&subsys->lock);
 	return ret;
diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c
index 8e0d766d2722..dc1f0f647189 100644
--- a/drivers/nvme/target/tcp.c
+++ b/drivers/nvme/target/tcp.c
@@ -94,7 +94,6 @@ struct nvmet_tcp_queue {
 	struct socket		*sock;
 	struct nvmet_tcp_port	*port;
 	struct work_struct	io_work;
-	int			cpu;
 	struct nvmet_cq		nvme_cq;
 	struct nvmet_sq		nvme_sq;
 
@@ -144,7 +143,6 @@ struct nvmet_tcp_port {
 	struct work_struct	accept_work;
 	struct nvmet_port	*nport;
 	struct sockaddr_storage addr;
-	int			last_cpu;
 	void (*data_ready)(struct sock *);
 };
 
@@ -219,6 +217,11 @@ static inline void nvmet_tcp_put_cmd(struct nvmet_tcp_cmd *cmd)
 	list_add_tail(&cmd->entry, &cmd->queue->free_list);
 }
 
+static inline int queue_cpu(struct nvmet_tcp_queue *queue)
+{
+	return queue->sock->sk->sk_incoming_cpu;
+}
+
 static inline u8 nvmet_tcp_hdgst_len(struct nvmet_tcp_queue *queue)
 {
 	return queue->hdr_digest ? NVME_TCP_DIGEST_LENGTH : 0;
@@ -506,7 +509,7 @@ static void nvmet_tcp_queue_response(struct nvmet_req *req)
 	struct nvmet_tcp_queue	*queue = cmd->queue;
 
 	llist_add(&cmd->lentry, &queue->resp_list);
-	queue_work_on(cmd->queue->cpu, nvmet_tcp_wq, &cmd->queue->io_work);
+	queue_work_on(queue_cpu(queue), nvmet_tcp_wq, &cmd->queue->io_work);
 }
 
 static int nvmet_try_send_data_pdu(struct nvmet_tcp_cmd *cmd)
@@ -1223,7 +1226,7 @@ static void nvmet_tcp_io_work(struct work_struct *w)
 	 * We exahusted our budget, requeue our selves
 	 */
 	if (pending)
-		queue_work_on(queue->cpu, nvmet_tcp_wq, &queue->io_work);
+		queue_work_on(queue_cpu(queue), nvmet_tcp_wq, &queue->io_work);
 }
 
 static int nvmet_tcp_alloc_cmd(struct nvmet_tcp_queue *queue,
@@ -1383,7 +1386,7 @@ static void nvmet_tcp_data_ready(struct sock *sk)
 	read_lock_bh(&sk->sk_callback_lock);
 	queue = sk->sk_user_data;
 	if (likely(queue))
-		queue_work_on(queue->cpu, nvmet_tcp_wq, &queue->io_work);
+		queue_work_on(queue_cpu(queue), nvmet_tcp_wq, &queue->io_work);
 	read_unlock_bh(&sk->sk_callback_lock);
 }
 
@@ -1403,7 +1406,7 @@ static void nvmet_tcp_write_space(struct sock *sk)
 
 	if (sk_stream_is_writeable(sk)) {
 		clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
-		queue_work_on(queue->cpu, nvmet_tcp_wq, &queue->io_work);
+		queue_work_on(queue_cpu(queue), nvmet_tcp_wq, &queue->io_work);
 	}
 out:
 	read_unlock_bh(&sk->sk_callback_lock);
@@ -1512,9 +1515,6 @@ static int nvmet_tcp_alloc_queue(struct nvmet_tcp_port *port,
 	if (ret)
 		goto out_free_connect;
 
-	port->last_cpu = cpumask_next_wrap(port->last_cpu,
-				cpu_online_mask, -1, false);
-	queue->cpu = port->last_cpu;
 	nvmet_prepare_receive_pdu(queue);
 
 	mutex_lock(&nvmet_tcp_queue_mutex);
@@ -1525,7 +1525,7 @@ static int nvmet_tcp_alloc_queue(struct nvmet_tcp_port *port,
 	if (ret)
 		goto out_destroy_sq;
 
-	queue_work_on(queue->cpu, nvmet_tcp_wq, &queue->io_work);
+	queue_work_on(queue_cpu(queue), nvmet_tcp_wq, &queue->io_work);
 
 	return 0;
 out_destroy_sq:
@@ -1612,7 +1612,6 @@ static int nvmet_tcp_add_port(struct nvmet_port *nport)
 	}
 
 	port->nport = nport;
-	port->last_cpu = -1;
 	INIT_WORK(&port->accept_work, nvmet_tcp_accept_work);
 	if (port->nport->inline_data_size < 0)
 		port->nport->inline_data_size = NVMET_TCP_DEF_INLINE_DATA_SIZE;
diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index 2bdd64648ef0..e6acbf940712 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -8,6 +8,8 @@
 #define _HISI_SAS_H_
 
 #include <linux/acpi.h>
+#include <linux/blk-mq.h>
+#include <linux/blk-mq-pci.h>
 #include <linux/clk.h>
 #include <linux/debugfs.h>
 #include <linux/dmapool.h>
@@ -431,7 +433,6 @@ struct hisi_hba {
 	u32 intr_coal_count;	/* Interrupt count to coalesce */
 
 	int cq_nvecs;
-	unsigned int *reply_map;
 
 	/* bist */
 	enum sas_linkrate debugfs_bist_linkrate;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index d9d21d23372e..a994c7b8d26f 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -417,6 +417,7 @@ static int hisi_sas_task_prep(struct sas_task *task,
 	struct device *dev = hisi_hba->dev;
 	int dlvry_queue_slot, dlvry_queue, rc, slot_idx;
 	int n_elem = 0, n_elem_dif = 0, n_elem_req = 0;
+	struct scsi_cmnd *scmd = NULL;
 	struct hisi_sas_dq *dq;
 	unsigned long flags;
 	int wr_q_index;
@@ -432,10 +433,23 @@ static int hisi_sas_task_prep(struct sas_task *task,
 		return -ECOMM;
 	}
 
-	if (hisi_hba->reply_map) {
-		int cpu = raw_smp_processor_id();
-		unsigned int dq_index = hisi_hba->reply_map[cpu];
+	if (task->uldd_task) {
+		struct ata_queued_cmd *qc;
 
+		if (dev_is_sata(device)) {
+			qc = task->uldd_task;
+			scmd = qc->scsicmd;
+		} else {
+			scmd = task->uldd_task;
+		}
+	}
+
+	if (scmd) {
+		unsigned int dq_index;
+		u32 blk_tag;
+
+		blk_tag = blk_mq_unique_tag(scmd->request);
+		dq_index = blk_mq_unique_tag_to_hwq(blk_tag);
 		*dq_pointer = dq = &hisi_hba->dq[dq_index];
 	} else {
 		*dq_pointer = dq = sas_dev->dq;
@@ -464,21 +478,9 @@ static int hisi_sas_task_prep(struct sas_task *task,
 
 	if (hisi_hba->hw->slot_index_alloc)
 		rc = hisi_hba->hw->slot_index_alloc(hisi_hba, device);
-	else {
-		struct scsi_cmnd *scsi_cmnd = NULL;
-
-		if (task->uldd_task) {
-			struct ata_queued_cmd *qc;
+	else
+		rc = hisi_sas_slot_index_alloc(hisi_hba, scmd);
 
-			if (dev_is_sata(device)) {
-				qc = task->uldd_task;
-				scsi_cmnd = qc->scsicmd;
-			} else {
-				scsi_cmnd = task->uldd_task;
-			}
-		}
-		rc  = hisi_sas_slot_index_alloc(hisi_hba, scsi_cmnd);
-	}
 	if (rc < 0)
 		goto err_out_dif_dma_unmap;
 
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
index 60adf5c32143..b6f75a1764df 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
@@ -2362,68 +2362,36 @@ static irqreturn_t cq_interrupt_v3_hw(int irq_no, void *p)
 	return IRQ_WAKE_THREAD;
 }
 
-static void setup_reply_map_v3_hw(struct hisi_hba *hisi_hba, int nvecs)
+static int interrupt_preinit_v3_hw(struct hisi_hba *hisi_hba)
 {
-	const struct cpumask *mask;
-	int queue, cpu;
+	int vectors;
+	int max_msi = HISI_SAS_MSI_COUNT_V3_HW, min_msi;
+	struct Scsi_Host *shost = hisi_hba->shost;
+	struct irq_affinity desc = {
+		.pre_vectors = BASE_VECTORS_V3_HW,
+	};
 
-	for (queue = 0; queue < nvecs; queue++) {
-		struct hisi_sas_cq *cq = &hisi_hba->cq[queue];
+	min_msi = MIN_AFFINE_VECTORS_V3_HW;
+	vectors = pci_alloc_irq_vectors_affinity(hisi_hba->pci_dev,
+						 min_msi, max_msi,
+						 PCI_IRQ_MSI |
+						 PCI_IRQ_AFFINITY,
+						 &desc);
+	if (vectors < 0)
+		return -ENOENT;
 
-		mask = pci_irq_get_affinity(hisi_hba->pci_dev, queue +
-					    BASE_VECTORS_V3_HW);
-		if (!mask)
-			goto fallback;
-		cq->irq_mask = mask;
-		for_each_cpu(cpu, mask)
-			hisi_hba->reply_map[cpu] = queue;
-	}
-	return;
 
-fallback:
-	for_each_possible_cpu(cpu)
-		hisi_hba->reply_map[cpu] = cpu % hisi_hba->queue_count;
-	/* Don't clean all CQ masks */
+	hisi_hba->cq_nvecs = vectors - BASE_VECTORS_V3_HW;
+	shost->nr_hw_queues = hisi_hba->cq_nvecs;
+
+	return 0;
 }
 
 static int interrupt_init_v3_hw(struct hisi_hba *hisi_hba)
 {
 	struct device *dev = hisi_hba->dev;
 	struct pci_dev *pdev = hisi_hba->pci_dev;
-	int vectors, rc, i;
-	int max_msi = HISI_SAS_MSI_COUNT_V3_HW, min_msi;
-
-	if (auto_affine_msi_experimental) {
-		struct irq_affinity desc = {
-			.pre_vectors = BASE_VECTORS_V3_HW,
-		};
-
-		dev_info(dev, "Enable MSI auto-affinity\n");
-
-		min_msi = MIN_AFFINE_VECTORS_V3_HW;
-
-		hisi_hba->reply_map = devm_kcalloc(dev, nr_cpu_ids,
-						   sizeof(unsigned int),
-						   GFP_KERNEL);
-		if (!hisi_hba->reply_map)
-			return -ENOMEM;
-		vectors = pci_alloc_irq_vectors_affinity(hisi_hba->pci_dev,
-							 min_msi, max_msi,
-							 PCI_IRQ_MSI |
-							 PCI_IRQ_AFFINITY,
-							 &desc);
-		if (vectors < 0)
-			return -ENOENT;
-		setup_reply_map_v3_hw(hisi_hba, vectors - BASE_VECTORS_V3_HW);
-	} else {
-		min_msi = max_msi;
-		vectors = pci_alloc_irq_vectors(hisi_hba->pci_dev, min_msi,
-						max_msi, PCI_IRQ_MSI);
-		if (vectors < 0)
-			return vectors;
-	}
-
-	hisi_hba->cq_nvecs = vectors - BASE_VECTORS_V3_HW;
+	int rc, i;
 
 	rc = devm_request_irq(dev, pci_irq_vector(pdev, 1),
 			      int_phy_up_down_bcast_v3_hw, 0,
@@ -3072,6 +3040,15 @@ static int debugfs_set_bist_v3_hw(struct hisi_hba *hisi_hba, bool enable)
 	return 0;
 }
 
+static int hisi_sas_map_queues(struct Scsi_Host *shost)
+{
+	struct hisi_hba *hisi_hba = shost_priv(shost);
+	struct blk_mq_queue_map *qmap = &shost->tag_set.map[HCTX_TYPE_DEFAULT];
+
+	return blk_mq_pci_map_queues(qmap, hisi_hba->pci_dev,
+				     BASE_VECTORS_V3_HW);
+}
+
 static struct scsi_host_template sht_v3_hw = {
 	.name			= DRV_NAME,
 	.proc_name		= DRV_NAME,
@@ -3082,6 +3059,7 @@ static struct scsi_host_template sht_v3_hw = {
 	.slave_configure	= hisi_sas_slave_configure,
 	.scan_finished		= hisi_sas_scan_finished,
 	.scan_start		= hisi_sas_scan_start,
+	.map_queues		= hisi_sas_map_queues,
 	.change_queue_depth	= sas_change_queue_depth,
 	.bios_param		= sas_bios_param,
 	.this_id		= -1,
@@ -3098,6 +3076,7 @@ static struct scsi_host_template sht_v3_hw = {
 	.shost_attrs		= host_attrs_v3_hw,
 	.tag_alloc_policy	= BLK_TAG_ALLOC_RR,
 	.host_reset             = hisi_sas_host_reset,
+	.host_tagset		= 1,
 };
 
 static const struct hisi_sas_hw hisi_sas_v3_hw = {
@@ -3269,6 +3248,10 @@ hisi_sas_v3_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (hisi_sas_debugfs_enable)
 		hisi_sas_debugfs_init(hisi_hba);
 
+	rc = interrupt_preinit_v3_hw(hisi_hba);
+	if (rc)
+		goto err_out_ha;
+	dev_err(dev, "%d hw queues\n", shost->nr_hw_queues);
 	rc = scsi_add_host(shost, dev);
 	if (rc)
 		goto err_out_ha;
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 37d1c5565d90..2f162603876f 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -421,6 +421,7 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize)
 	shost->cmd_per_lun = sht->cmd_per_lun;
 	shost->unchecked_isa_dma = sht->unchecked_isa_dma;
 	shost->no_write_same = sht->no_write_same;
+	shost->host_tagset = sht->host_tagset;
 
 	if (shost_eh_deadline == -1 || !sht->eh_host_reset_handler)
 		shost->eh_deadline = -1;
diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c
index 2b7e7b5f38ed..020270ce790b 100644
--- a/drivers/scsi/megaraid/megaraid_sas_base.c
+++ b/drivers/scsi/megaraid/megaraid_sas_base.c
@@ -37,6 +37,7 @@
 #include <linux/poll.h>
 #include <linux/vmalloc.h>
 #include <linux/irq_poll.h>
+#include <linux/blk-mq-pci.h>
 
 #include <scsi/scsi.h>
 #include <scsi/scsi_cmnd.h>
@@ -113,6 +114,10 @@ unsigned int enable_sdev_max_qd;
 module_param(enable_sdev_max_qd, int, 0444);
 MODULE_PARM_DESC(enable_sdev_max_qd, "Enable sdev max qd as can_queue. Default: 0");
 
+int host_tagset_enable = 1;
+module_param(host_tagset_enable, int, 0444);
+MODULE_PARM_DESC(host_tagset_enable, "Shared host tagset enable/disable Default: enable(1)");
+
 MODULE_LICENSE("GPL");
 MODULE_VERSION(MEGASAS_VERSION);
 MODULE_AUTHOR("megaraidlinux.pdl@broadcom.com");
@@ -3119,6 +3124,19 @@ megasas_bios_param(struct scsi_device *sdev, struct block_device *bdev,
 	return 0;
 }
 
+static int megasas_map_queues(struct Scsi_Host *shost)
+{
+	struct megasas_instance *instance;
+
+	instance = (struct megasas_instance *)shost->hostdata;
+
+	if (shost->nr_hw_queues == 1)
+		return 0;
+
+	return blk_mq_pci_map_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT],
+			instance->pdev, instance->low_latency_index_start);
+}
+
 static void megasas_aen_polling(struct work_struct *work);
 
 /**
@@ -3427,6 +3445,7 @@ static struct scsi_host_template megasas_template = {
 	.eh_timed_out = megasas_reset_timer,
 	.shost_attrs = megaraid_host_attrs,
 	.bios_param = megasas_bios_param,
+	.map_queues = megasas_map_queues,
 	.change_queue_depth = scsi_change_queue_depth,
 	.max_segment_size = 0xffffffff,
 };
@@ -6808,6 +6827,26 @@ static int megasas_io_attach(struct megasas_instance *instance)
 	host->max_lun = MEGASAS_MAX_LUN;
 	host->max_cmd_len = 16;
 
+	/* Use shared host tagset only for fusion adaptors
+	 * if there are managed interrupts (smp affinity enabled case).
+	 * Single msix_vectors in kdump, so shared host tag is also disabled.
+	 */
+
+	host->host_tagset = 0;
+	host->nr_hw_queues = 1;
+
+	if ((instance->adapter_type != MFI_SERIES) &&
+		(instance->msix_vectors > instance->low_latency_index_start) &&
+		host_tagset_enable &&
+		instance->smp_affinity_enable) {
+		host->host_tagset = 1;
+		host->nr_hw_queues = instance->msix_vectors -
+			instance->low_latency_index_start;
+	}
+
+	dev_info(&instance->pdev->dev,
+		"Max firmware commands: %d shared with nr_hw_queues = %d\n",
+		instance->max_fw_cmds, host->nr_hw_queues);
 	/*
 	 * Notify the mid-layer about the new controller
 	 */
diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c
index b0c01cf0428f..fd607287608e 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -359,24 +359,29 @@ megasas_get_msix_index(struct megasas_instance *instance,
 {
 	int sdev_busy;
 
-	/* nr_hw_queue = 1 for MegaRAID */
-	struct blk_mq_hw_ctx *hctx =
-		scmd->device->request_queue->queue_hw_ctx[0];
-
-	sdev_busy = atomic_read(&hctx->nr_active);
+	/* TBD - if sml remove device_busy in future, driver
+	 * should track counter in internal structure.
+	 */
+	sdev_busy = atomic_read(&scmd->device->device_busy);
 
 	if (instance->perf_mode == MR_BALANCED_PERF_MODE &&
-	    sdev_busy > (data_arms * MR_DEVICE_HIGH_IOPS_DEPTH))
+	    sdev_busy > (data_arms * MR_DEVICE_HIGH_IOPS_DEPTH)) {
 		cmd->request_desc->SCSIIO.MSIxIndex =
 			mega_mod64((atomic64_add_return(1, &instance->high_iops_outstanding) /
 					MR_HIGH_IOPS_BATCH_COUNT), instance->low_latency_index_start);
-	else if (instance->msix_load_balance)
+	} else if (instance->msix_load_balance) {
 		cmd->request_desc->SCSIIO.MSIxIndex =
 			(mega_mod64(atomic64_add_return(1, &instance->total_io_count),
 				instance->msix_vectors));
-	else
+	} else if (instance->host->nr_hw_queues > 1) {
+		u32 tag = blk_mq_unique_tag(scmd->request);
+
+		cmd->request_desc->SCSIIO.MSIxIndex = blk_mq_unique_tag_to_hwq(tag) +
+			instance->low_latency_index_start;
+	} else {
 		cmd->request_desc->SCSIIO.MSIxIndex =
 			instance->reply_map[raw_smp_processor_id()];
+	}
 }
 
 /**
@@ -956,9 +961,6 @@ megasas_alloc_cmds_fusion(struct megasas_instance *instance)
 	if (megasas_alloc_cmdlist_fusion(instance))
 		goto fail_exit;
 
-	dev_info(&instance->pdev->dev, "Configured max firmware commands: %d\n",
-		 instance->max_fw_cmds);
-
 	/* The first 256 bytes (SMID 0) is not used. Don't add to the cmd list */
 	io_req_base = fusion->io_request_frames + MEGA_MPI2_RAID_DEFAULT_IO_FRAME_SIZE;
 	io_req_base_phys = fusion->io_request_frames_phys + MEGA_MPI2_RAID_DEFAULT_IO_FRAME_SIZE;
@@ -1102,8 +1104,9 @@ megasas_ioc_init_fusion(struct megasas_instance *instance)
 		MR_HIGH_IOPS_QUEUE_COUNT) && cur_intr_coalescing)
 		instance->perf_mode = MR_BALANCED_PERF_MODE;
 
-	dev_info(&instance->pdev->dev, "Performance mode :%s\n",
-		MEGASAS_PERF_MODE_2STR(instance->perf_mode));
+	dev_info(&instance->pdev->dev, "Performance mode :%s (latency index = %d)\n",
+		MEGASAS_PERF_MODE_2STR(instance->perf_mode),
+		instance->low_latency_index_start);
 
 	instance->fw_sync_cache_support = (scratch_pad_1 &
 		MR_CAN_HANDLE_SYNC_CACHE_OFFSET) ? 1 : 0;
diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index 1ad7260d4758..a87e40aec11f 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -4696,19 +4696,14 @@ fini:
 static struct sdebug_queue *get_queue(struct scsi_cmnd *cmnd)
 {
 	u16 hwq;
+	u32 tag = blk_mq_unique_tag(cmnd->request);
 
-	if (sdebug_host_max_queue) {
-		/* Provide a simple method to choose the hwq */
-		hwq = smp_processor_id() % submit_queues;
-	} else {
-		u32 tag = blk_mq_unique_tag(cmnd->request);
+	hwq = blk_mq_unique_tag_to_hwq(tag);
 
-		hwq = blk_mq_unique_tag_to_hwq(tag);
+	pr_debug("tag=%#x, hwq=%d\n", tag, hwq);
+	if (WARN_ON_ONCE(hwq >= submit_queues))
+		hwq = 0;
 
-		pr_debug("tag=%#x, hwq=%d\n", tag, hwq);
-		if (WARN_ON_ONCE(hwq >= submit_queues))
-			hwq = 0;
-	}
 	return sdebug_q_arr + hwq;
 }
 
@@ -7347,10 +7342,7 @@ static int sdebug_driver_probe(struct device *dev)
 
 	sdbg_host = to_sdebug_host(dev);
 
-	if (sdebug_host_max_queue)
-		sdebug_driver_template.can_queue = sdebug_host_max_queue;
-	else
-		sdebug_driver_template.can_queue = sdebug_max_queue;
+	sdebug_driver_template.can_queue = sdebug_max_queue;
 	if (!sdebug_clustering)
 		sdebug_driver_template.dma_boundary = PAGE_SIZE - 1;
 
@@ -7367,11 +7359,11 @@ static int sdebug_driver_probe(struct device *dev)
 	}
 	/*
 	 * Decide whether to tell scsi subsystem that we want mq. The
-	 * following should give the same answer for each host. If the host
-	 * has a limit of hostwide max commands, then do not set.
+	 * following should give the same answer for each host.
 	 */
-	if (!sdebug_host_max_queue)
-		hpnt->nr_hw_queues = submit_queues;
+	hpnt->nr_hw_queues = submit_queues;
+	if (sdebug_host_max_queue)
+		hpnt->host_tagset = 1;
 
 	sdbg_host->shost = hpnt;
 	*((struct sdebug_host_info **)hpnt->hostdata) = sdbg_host;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 7affaaf8b98e..a89478a0c588 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1891,6 +1891,8 @@ int scsi_mq_setup_tags(struct Scsi_Host *shost)
 	tag_set->flags |=
 		BLK_ALLOC_POLICY_TO_MQ_FLAG(shost->hostt->tag_alloc_policy);
 	tag_set->driver_data = shost;
+	if (shost->host_tagset)
+		tag_set->flags |= BLK_MQ_F_TAG_HCTX_SHARED;
 
 	return blk_mq_alloc_tag_set(tag_set);
 }
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index 163dbcb741c1..d6e344fa33ad 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -393,6 +393,16 @@ show_use_blk_mq(struct device *dev, struct device_attribute *attr, char *buf)
 }
 static DEVICE_ATTR(use_blk_mq, S_IRUGO, show_use_blk_mq, NULL);
 
+static ssize_t
+show_nr_hw_queues(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct Scsi_Host *shost = class_to_shost(dev);
+	struct blk_mq_tag_set *tag_set = &shost->tag_set;
+
+	return snprintf(buf, 20, "%d\n", tag_set->nr_hw_queues);
+}
+static DEVICE_ATTR(nr_hw_queues, S_IRUGO, show_nr_hw_queues, NULL);
+
 static struct attribute *scsi_sysfs_shost_attrs[] = {
 	&dev_attr_use_blk_mq.attr,
 	&dev_attr_unique_id.attr,
@@ -411,6 +421,7 @@ static struct attribute *scsi_sysfs_shost_attrs[] = {
 	&dev_attr_prot_guard_type.attr,
 	&dev_attr_host_reset.attr,
 	&dev_attr_eh_deadline.attr,
+	&dev_attr_nr_hw_queues.attr,
 	NULL
 };