Merge tag 'dm-3.20-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper changes from Mike Snitzer:

- The most significant change this cycle is request-based DM now
supports stacking ontop of blk-mq devices. This blk-mq support
changes the model request-based DM uses for cloning a request to
relying on calling blk_get_request() directly from the underlying
blk-mq device.

An early consumer of this code is Intel's emerging NVMe hardware;
thanks to Keith Busch for working on, and pushing for, these changes.

- A few other small fixes and cleanups across other DM targets.

* tag 'dm-3.20-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm: inherit QUEUE_FLAG_SG_GAPS flags from underlying queues
dm snapshot: remove unnecessary NULL checks before vfree() calls
dm mpath: simplify failure path of dm_multipath_init()
dm thin metadata: remove unused dm_pool_get_data_block_size()
dm ioctl: fix stale comment above dm_get_inactive_table()
dm crypt: update url in CONFIG_DM_CRYPT help text
dm bufio: fix time comparison to use time_after_eq()
dm: use time_in_range() and time_after()
dm raid: fix a couple integer overflows
dm table: train hybrid target type detection to select blk-mq if appropriate
dm: allocate requests in target when stacking on blk-mq devices
dm: prepare for allocating blk-mq clone requests in target
dm: submit stacked requests in irq enabled context
dm: split request structure out from dm_rq_target_io structure
dm: remove exports for request-based interfaces without external callers

drivers/md/Kconfig

···
 	  transparently encrypts the data on it. You'll need to activate
 	  the ciphers you're going to use in the cryptoapi configuration.
 
-	  Information on how to use dm-crypt can be found on
-
-	  <http://www.saout.de/misc/dm-crypt/>
+	  For further information on dm-crypt and userspace tools see:
+	  <http://code.google.com/p/cryptsetup/wiki/DMCrypt>
 
 	  To compile this code as a module, choose M here: the module will
 	  be called dm-crypt.

drivers/md/dm-bufio.c

···
 #include <linux/device-mapper.h>
 #include <linux/dm-io.h>
 #include <linux/slab.h>
+#include <linux/jiffies.h>
 #include <linux/vmalloc.h>
 #include <linux/shrinker.h>
 #include <linux/module.h>
···
 
 static bool older_than(struct dm_buffer *b, unsigned long age_hz)
 {
-	return (jiffies - b->last_accessed) >= age_hz;
+	return time_after_eq(jiffies, b->last_accessed + age_hz);
 }
 
 static void __evict_old_buffers(struct dm_bufio_client *c, unsigned long age_hz)

drivers/md/dm-cache-target.c

···
 
 #include <linux/dm-io.h>
 #include <linux/dm-kcopyd.h>
+#include <linux/jiffies.h>
 #include <linux/init.h>
 #include <linux/mempool.h>
 #include <linux/module.h>
···
 
 static int need_commit_due_to_time(struct cache *cache)
 {
-	return jiffies < cache->last_commit_jiffies ||
-	       jiffies > cache->last_commit_jiffies + COMMIT_PERIOD;
+	return !time_in_range(jiffies, cache->last_commit_jiffies,
+			      cache->last_commit_jiffies + COMMIT_PERIOD);
 }
 
 static int commit_if_needed(struct cache *cache)

drivers/md/dm-ioctl.c

···
 
 /*
  * On successful return, the caller must not attempt to acquire
- * _hash_lock without first calling dm_table_put, because dm_table_destroy
- * waits for this dm_table_put and could be called under this lock.
+ * _hash_lock without first calling dm_put_live_table, because dm_table_destroy
+ * waits for this dm_put_live_table and could be called under this lock.
  */
 static struct dm_table *dm_get_inactive_table(struct mapped_device *md, int *srcu_idx)
 {

drivers/md/dm-log-userspace-base.c

···
 
 #include <linux/bio.h>
 #include <linux/slab.h>
+#include <linux/jiffies.h>
 #include <linux/dm-dirty-log.h>
 #include <linux/device-mapper.h>
 #include <linux/dm-log-userspace.h>
···
 	int r;
 	uint64_t region64 = region;
 	struct log_c *lc = log->context;
-	static unsigned long long limit;
+	static unsigned long limit;
 	struct {
 		int64_t is_recovering;
 		uint64_t in_sync_hint;
···
 	 */
 	if (region < lc->in_sync_hint)
 		return 0;
-	else if (jiffies < limit)
+	else if (time_after(limit, jiffies))
 		return 1;
 
 	limit = jiffies + (HZ / 4);

drivers/md/dm-mpath.c

···
 #include "dm-path-selector.h"
 #include "dm-uevent.h"
 
+#include <linux/blkdev.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
 #include <linux/mempool.h>
···
 /*
  * Map cloned requests
  */
-static int multipath_map(struct dm_target *ti, struct request *clone,
-			 union map_info *map_context)
+static int __multipath_map(struct dm_target *ti, struct request *clone,
+			   union map_info *map_context,
+			   struct request *rq, struct request **__clone)
 {
 	struct multipath *m = (struct multipath *) ti->private;
 	int r = DM_MAPIO_REQUEUE;
-	size_t nr_bytes = blk_rq_bytes(clone);
-	unsigned long flags;
+	size_t nr_bytes = clone ? blk_rq_bytes(clone) : blk_rq_bytes(rq);
 	struct pgpath *pgpath;
 	struct block_device *bdev;
 	struct dm_mpath_io *mpio;
 
-	spin_lock_irqsave(&m->lock, flags);
+	spin_lock_irq(&m->lock);
 
 	/* Do we need to select a new pgpath? */
 	if (!m->current_pgpath ||
···
 		/* ENOMEM, requeue */
 		goto out_unlock;
 
-	bdev = pgpath->path.dev->bdev;
-	clone->q = bdev_get_queue(bdev);
-	clone->rq_disk = bdev->bd_disk;
-	clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
 	mpio = map_context->ptr;
 	mpio->pgpath = pgpath;
 	mpio->nr_bytes = nr_bytes;
+
+	bdev = pgpath->path.dev->bdev;
+
+	spin_unlock_irq(&m->lock);
+
+	if (clone) {
+		/* Old request-based interface: allocated clone is passed in */
+		clone->q = bdev_get_queue(bdev);
+		clone->rq_disk = bdev->bd_disk;
+		clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
+	} else {
+		/* blk-mq request-based interface */
+		*__clone = blk_get_request(bdev_get_queue(bdev),
+					   rq_data_dir(rq), GFP_KERNEL);
+		if (IS_ERR(*__clone))
+			/* ENOMEM, requeue */
+			return r;
+		(*__clone)->bio = (*__clone)->biotail = NULL;
+		(*__clone)->rq_disk = bdev->bd_disk;
+		(*__clone)->cmd_flags |= REQ_FAILFAST_TRANSPORT;
+	}
+
 	if (pgpath->pg->ps.type->start_io)
 		pgpath->pg->ps.type->start_io(&pgpath->pg->ps,
 					      &pgpath->path,
 					      nr_bytes);
-	r = DM_MAPIO_REMAPPED;
+	return DM_MAPIO_REMAPPED;
 
 out_unlock:
-	spin_unlock_irqrestore(&m->lock, flags);
+	spin_unlock_irq(&m->lock);
 
 	return r;
+}
+
+static int multipath_map(struct dm_target *ti, struct request *clone,
+			 union map_info *map_context)
+{
+	return __multipath_map(ti, clone, map_context, NULL, NULL);
+}
+
+static int multipath_clone_and_map(struct dm_target *ti, struct request *rq,
+				   union map_info *map_context,
+				   struct request **clone)
+{
+	return __multipath_map(ti, NULL, map_context, rq, clone);
+}
+
+static void multipath_release_clone(struct request *clone)
+{
+	blk_put_request(clone);
 }
 
 /*
···
  *---------------------------------------------------------------*/
 static struct target_type multipath_target = {
 	.name = "multipath",
-	.version = {1, 7, 0},
+	.version = {1, 8, 0},
 	.module = THIS_MODULE,
 	.ctr = multipath_ctr,
 	.dtr = multipath_dtr,
 	.map_rq = multipath_map,
+	.clone_and_map_rq = multipath_clone_and_map,
+	.release_clone_rq = multipath_release_clone,
 	.rq_end_io = multipath_end_io,
 	.presuspend = multipath_presuspend,
 	.postsuspend = multipath_postsuspend,
···
 	r = dm_register_target(&multipath_target);
 	if (r < 0) {
 		DMERR("register failed %d", r);
-		kmem_cache_destroy(_mpio_cache);
-		return -EINVAL;
+		r = -EINVAL;
+		goto bad_register_target;
 	}
 
 	kmultipathd = alloc_workqueue("kmpathd", WQ_MEM_RECLAIM, 0);
 	if (!kmultipathd) {
 		DMERR("failed to create workqueue kmpathd");
-		dm_unregister_target(&multipath_target);
-		kmem_cache_destroy(_mpio_cache);
-		return -ENOMEM;
+		r = -ENOMEM;
+		goto bad_alloc_kmultipathd;
 	}
 
 	/*
···
 						  WQ_MEM_RECLAIM);
 	if (!kmpath_handlerd) {
 		DMERR("failed to create workqueue kmpath_handlerd");
-		destroy_workqueue(kmultipathd);
-		dm_unregister_target(&multipath_target);
-		kmem_cache_destroy(_mpio_cache);
-		return -ENOMEM;
+		r = -ENOMEM;
+		goto bad_alloc_kmpath_handlerd;
 	}
 
 	DMINFO("version %u.%u.%u loaded",
 	       multipath_target.version[0], multipath_target.version[1],
 	       multipath_target.version[2]);
+
+	return 0;
+
+bad_alloc_kmpath_handlerd:
+	destroy_workqueue(kmultipathd);
+bad_alloc_kmultipathd:
+	dm_unregister_target(&multipath_target);
+bad_register_target:
+	kmem_cache_destroy(_mpio_cache);
 
 	return r;
 }

drivers/md/dm-raid.c

···
 	argv++;
 
 	/* Skip over RAID params for now and find out # of devices */
-	if (num_raid_params + 1 > argc) {
+	if (num_raid_params >= argc) {
 		ti->error = "Arguments do not agree with counts given";
 		return -EINVAL;
 	}
···
 	if ((kstrtoul(argv[num_raid_params], 10, &num_raid_devs) < 0) ||
 	    (num_raid_devs >= INT_MAX)) {
 		ti->error = "Cannot understand number of raid devices";
+		return -EINVAL;
+	}
+
+	argc -= num_raid_params + 1; /* +1: we already have num_raid_devs */
+	if (argc != (num_raid_devs * 2)) {
+		ti->error = "Supplied RAID devices does not match the count given";
 		return -EINVAL;
 	}
 
···
 	if (ret)
 		goto bad;
 
-	ret = -EINVAL;
-
-	argc -= num_raid_params + 1; /* +1: we already have num_raid_devs */
 	argv += num_raid_params + 1;
-
-	if (argc != (num_raid_devs * 2)) {
-		ti->error = "Supplied RAID devices does not match the count given";
-		goto bad;
-	}
 
 	ret = dev_parms(rs, argv);
 	if (ret)

drivers/md/dm-snap-persistent.c

···
 
 static void free_area(struct pstore *ps)
 {
-	if (ps->area)
-		vfree(ps->area);
+	vfree(ps->area);
 	ps->area = NULL;
-
-	if (ps->zero_area)
-		vfree(ps->zero_area);
+	vfree(ps->zero_area);
 	ps->zero_area = NULL;
-
-	if (ps->header_area)
-		vfree(ps->header_area);
+	vfree(ps->header_area);
 	ps->header_area = NULL;
 }
 
···
 	free_area(ps);
 
 	/* Allocated in persistent_read_metadata */
-	if (ps->callbacks)
-		vfree(ps->callbacks);
+	vfree(ps->callbacks);
 
 	kfree(ps);
 }

drivers/md/dm-table.c

···
 {
 	unsigned i;
 	unsigned bio_based = 0, request_based = 0, hybrid = 0;
+	bool use_blk_mq = false;
 	struct dm_target *tgt;
 	struct dm_dev_internal *dd;
 	struct list_head *devices;
-	unsigned live_md_type;
+	unsigned live_md_type = dm_get_md_type(t->md);
 
 	for (i = 0; i < t->num_targets; i++) {
 		tgt = t->targets + i;
···
 		 * Determine the type from the live device.
 		 * Default to bio-based if device is new.
 		 */
-		live_md_type = dm_get_md_type(t->md);
-		if (live_md_type == DM_TYPE_REQUEST_BASED)
+		if (live_md_type == DM_TYPE_REQUEST_BASED ||
+		    live_md_type == DM_TYPE_MQ_REQUEST_BASED)
 			request_based = 1;
 		else
 			bio_based = 1;
···
 
 	BUG_ON(!request_based); /* No targets in this table */
 
-	/* Non-request-stackable devices can't be used for request-based dm */
-	devices = dm_table_get_devices(t);
-	list_for_each_entry(dd, devices, list) {
-		if (!blk_queue_stackable(bdev_get_queue(dd->dm_dev->bdev))) {
-			DMWARN("table load rejected: including"
-			       " non-request-stackable devices");
-			return -EINVAL;
-		}
-	}
-
 	/*
 	 * Request-based dm supports only tables that have a single target now.
 	 * To support multiple targets, request splitting support is needed,
···
 		return -EINVAL;
 	}
 
-	t->type = DM_TYPE_REQUEST_BASED;
+	/* Non-request-stackable devices can't be used for request-based dm */
+	devices = dm_table_get_devices(t);
+	list_for_each_entry(dd, devices, list) {
+		struct request_queue *q = bdev_get_queue(dd->dm_dev->bdev);
+
+		if (!blk_queue_stackable(q)) {
+			DMERR("table load rejected: including"
+			      " non-request-stackable devices");
+			return -EINVAL;
+		}
+
+		if (q->mq_ops)
+			use_blk_mq = true;
+	}
+
+	if (use_blk_mq) {
+		/* verify _all_ devices in the table are blk-mq devices */
+		list_for_each_entry(dd, devices, list)
+			if (!bdev_get_queue(dd->dm_dev->bdev)->mq_ops) {
+				DMERR("table load rejected: not all devices"
+				      " are blk-mq request-stackable");
+				return -EINVAL;
+			}
+		t->type = DM_TYPE_MQ_REQUEST_BASED;
+
+	} else if (hybrid && list_empty(devices) && live_md_type != DM_TYPE_NONE) {
+		/* inherit live MD type */
+		t->type = live_md_type;
+
+	} else
+		t->type = DM_TYPE_REQUEST_BASED;
 
 	return 0;
 }
···
 
 bool dm_table_request_based(struct dm_table *t)
 {
-	return dm_table_get_type(t) == DM_TYPE_REQUEST_BASED;
+	unsigned table_type = dm_table_get_type(t);
+
+	return (table_type == DM_TYPE_REQUEST_BASED ||
+		table_type == DM_TYPE_MQ_REQUEST_BASED);
+}
+
+bool dm_table_mq_request_based(struct dm_table *t)
+{
+	return dm_table_get_type(t) == DM_TYPE_MQ_REQUEST_BASED;
 }
 
 static int dm_table_alloc_md_mempools(struct dm_table *t)
···
 	return q && !test_bit(QUEUE_FLAG_NO_SG_MERGE, &q->queue_flags);
 }
 
+static int queue_supports_sg_gaps(struct dm_target *ti, struct dm_dev *dev,
+				  sector_t start, sector_t len, void *data)
+{
+	struct request_queue *q = bdev_get_queue(dev->bdev);
+
+	return q && !test_bit(QUEUE_FLAG_SG_GAPS, &q->queue_flags);
+}
+
 static bool dm_table_all_devices_attribute(struct dm_table *t,
 					   iterate_devices_callout_fn func)
 {
···
 		queue_flag_clear_unlocked(QUEUE_FLAG_NO_SG_MERGE, q);
 	else
 		queue_flag_set_unlocked(QUEUE_FLAG_NO_SG_MERGE, q);
+
+	if (dm_table_all_devices_attribute(t, queue_supports_sg_gaps))
+		queue_flag_clear_unlocked(QUEUE_FLAG_SG_GAPS, q);
+	else
+		queue_flag_set_unlocked(QUEUE_FLAG_SG_GAPS, q);
 
 	dm_table_set_integrity(t);
 

drivers/md/dm-target.c

···
 	return -EIO;
 }
 
+static int io_err_clone_and_map_rq(struct dm_target *ti, struct request *rq,
+				   union map_info *map_context,
+				   struct request **clone)
+{
+	return -EIO;
+}
+
+static void io_err_release_clone_rq(struct request *clone)
+{
+}
+
 static struct target_type error_target = {
 	.name = "error",
-	.version = {1, 2, 0},
+	.version = {1, 3, 0},
 	.ctr  = io_err_ctr,
 	.dtr  = io_err_dtr,
 	.map  = io_err_map,
 	.map_rq = io_err_map_rq,
+	.clone_and_map_rq = io_err_clone_and_map_rq,
+	.release_clone_rq = io_err_release_clone_rq,
 };
 
 int __init dm_target_init(void)

drivers/md/dm-thin-metadata.c

···
 	return r;
 }
 
-int dm_pool_get_data_block_size(struct dm_pool_metadata *pmd, sector_t *result)
-{
-	down_read(&pmd->root_lock);
-	*result = pmd->data_block_size;
-	up_read(&pmd->root_lock);
-
-	return 0;
-}
-
 int dm_pool_get_data_dev_size(struct dm_pool_metadata *pmd, dm_block_t *result)
 {
 	int r = -EINVAL;

drivers/md/dm-thin-metadata.h

···
 int dm_pool_get_metadata_dev_size(struct dm_pool_metadata *pmd,
 				  dm_block_t *result);
 
-int dm_pool_get_data_block_size(struct dm_pool_metadata *pmd, sector_t *result);
-
 int dm_pool_get_data_dev_size(struct dm_pool_metadata *pmd, dm_block_t *result);
 
 int dm_pool_block_is_used(struct dm_pool_metadata *pmd, dm_block_t b, bool *result);

drivers/md/dm-thin.c

···
 #include <linux/device-mapper.h>
 #include <linux/dm-io.h>
 #include <linux/dm-kcopyd.h>
+#include <linux/jiffies.h>
 #include <linux/log2.h>
 #include <linux/list.h>
 #include <linux/rculist.h>
···
  */
 static int need_commit_due_to_time(struct pool *pool)
 {
-	return jiffies < pool->last_commit_jiffies ||
-	       jiffies > pool->last_commit_jiffies + COMMIT_PERIOD;
+	return !time_in_range(jiffies, pool->last_commit_jiffies,
+			      pool->last_commit_jiffies + COMMIT_PERIOD);
 }
 
 #define thin_pbd(node) rb_entry((node), struct dm_thin_endio_hook, rb_node)

drivers/md/dm.c

···
 #include <linux/hdreg.h>
 #include <linux/delay.h>
 #include <linux/wait.h>
+#include <linux/kthread.h>
 
 #include <trace/events/block.h>
 
···
 struct dm_rq_target_io {
 	struct mapped_device *md;
 	struct dm_target *ti;
-	struct request *orig, clone;
+	struct request *orig, *clone;
+	struct kthread_work work;
 	int error;
 	union map_info info;
 };
···
 	 * io objects are allocated from here.
 	 */
 	mempool_t *io_pool;
+	mempool_t *rq_pool;
 
 	struct bio_set *bs;
 
···
 	unsigned internal_suspend_count;
 
 	struct dm_stats stats;
+
+	struct kthread_worker kworker;
+	struct task_struct *kworker_task;
 };
 
 /*
···
  */
 struct dm_md_mempools {
 	mempool_t *io_pool;
+	mempool_t *rq_pool;
 	struct bio_set *bs;
 };
 
···
 #define RESERVED_MAX_IOS		1024
 static struct kmem_cache *_io_cache;
 static struct kmem_cache *_rq_tio_cache;
+static struct kmem_cache *_rq_cache;
 
 /*
  * Bio-based DM's mempools' reserved IOs set by the user.
···
 	if (!_rq_tio_cache)
 		goto out_free_io_cache;
 
+	_rq_cache = kmem_cache_create("dm_clone_request", sizeof(struct request),
+				      __alignof__(struct request), 0, NULL);
+	if (!_rq_cache)
+		goto out_free_rq_tio_cache;
+
 	r = dm_uevent_init();
 	if (r)
-		goto out_free_rq_tio_cache;
+		goto out_free_rq_cache;
 
 	deferred_remove_workqueue = alloc_workqueue("kdmremove", WQ_UNBOUND, 1);
 	if (!deferred_remove_workqueue) {
···
 	destroy_workqueue(deferred_remove_workqueue);
 out_uevent_exit:
 	dm_uevent_exit();
+out_free_rq_cache:
+	kmem_cache_destroy(_rq_cache);
 out_free_rq_tio_cache:
 	kmem_cache_destroy(_rq_tio_cache);
 out_free_io_cache:
···
 	flush_scheduled_work();
 	destroy_workqueue(deferred_remove_workqueue);
 
+	kmem_cache_destroy(_rq_cache);
 	kmem_cache_destroy(_rq_tio_cache);
 	kmem_cache_destroy(_io_cache);
 	unregister_blkdev(_major, _name);
···
 static void free_rq_tio(struct dm_rq_target_io *tio)
 {
 	mempool_free(tio, tio->md->io_pool);
+}
+
+static struct request *alloc_clone_request(struct mapped_device *md,
+					   gfp_t gfp_mask)
+{
+	return mempool_alloc(md->rq_pool, gfp_mask);
+}
+
+static void free_clone_request(struct mapped_device *md, struct request *rq)
+{
+	mempool_free(rq, md->rq_pool);
 }
 
 static int md_in_flight(struct mapped_device *md)
···
  * the md may be freed in dm_put() at the end of this function.
  * Or do dm_get() before calling this function and dm_put() later.
  */
-static void rq_completed(struct mapped_device *md, int rw, int run_queue)
+static void rq_completed(struct mapped_device *md, int rw, bool run_queue)
 {
 	atomic_dec(&md->pending[rw]);
 
···
 	struct dm_rq_target_io *tio = clone->end_io_data;
 
 	blk_rq_unprep_clone(clone);
+	if (clone->q && clone->q->mq_ops)
+		tio->ti->type->release_clone_rq(clone);
+	else
+		free_clone_request(tio->md, clone);
 	free_rq_tio(tio);
 }
 
 /*
  * Complete the clone and the original request.
- * Must be called without queue lock.
+ * Must be called without clone's queue lock held,
+ * see end_clone_request() for more details.
  */
 static void dm_end_request(struct request *clone, int error)
 {
···
 
 static void dm_unprep_request(struct request *rq)
 {
-	struct request *clone = rq->special;
+	struct dm_rq_target_io *tio = rq->special;
+	struct request *clone = tio->clone;
 
 	rq->special = NULL;
 	rq->cmd_flags &= ~REQ_DONTPREP;
 
-	free_rq_clone(clone);
+	if (clone)
+		free_rq_clone(clone);
 }
 
 /*
  * Requeue the original request of a clone.
  */
-void dm_requeue_unmapped_request(struct request *clone)
+static void dm_requeue_unmapped_original_request(struct mapped_device *md,
+						 struct request *rq)
 {
-	int rw = rq_data_dir(clone);
-	struct dm_rq_target_io *tio = clone->end_io_data;
-	struct mapped_device *md = tio->md;
-	struct request *rq = tio->orig;
+	int rw = rq_data_dir(rq);
 	struct request_queue *q = rq->q;
 	unsigned long flags;
 
···
 	blk_requeue_request(q, rq);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 
-	rq_completed(md, rw, 0);
+	rq_completed(md, rw, false);
 }
-EXPORT_SYMBOL_GPL(dm_requeue_unmapped_request);
+
+static void dm_requeue_unmapped_request(struct request *clone)
+{
+	struct dm_rq_target_io *tio = clone->end_io_data;
+
+	dm_requeue_unmapped_original_request(tio->md, tio->orig);
+}
 
 static void __stop_queue(struct request_queue *q)
 {
···
 static void dm_softirq_done(struct request *rq)
 {
 	bool mapped = true;
-	struct request *clone = rq->completion_data;
-	struct dm_rq_target_io *tio = clone->end_io_data;
+	struct dm_rq_target_io *tio = rq->special;
+	struct request *clone = tio->clone;
+
+	if (!clone) {
+		blk_end_request_all(rq, tio->error);
+		rq_completed(tio->md, rq_data_dir(rq), false);
+		free_rq_tio(tio);
+		return;
+	}
 
 	if (rq->cmd_flags & REQ_FAILED)
 		mapped = false;
···
  * Complete the clone and the original request with the error status
  * through softirq context.
  */
-static void dm_complete_request(struct request *clone, int error)
+static void dm_complete_request(struct request *rq, int error)
 {
-	struct dm_rq_target_io *tio = clone->end_io_data;
-	struct request *rq = tio->orig;
+	struct dm_rq_target_io *tio = rq->special;
 
 	tio->error = error;
-	rq->completion_data = clone;
 	blk_complete_request(rq);
 }
 
···
  * Complete the not-mapped clone and the original request with the error status
  * through softirq context.
  * Target's rq_end_io() function isn't called.
- * This may be used when the target's map_rq() function fails.
+ * This may be used when the target's map_rq() or clone_and_map_rq() functions fail.
  */
-void dm_kill_unmapped_request(struct request *clone, int error)
+static void dm_kill_unmapped_request(struct request *rq, int error)
 {
-	struct dm_rq_target_io *tio = clone->end_io_data;
-	struct request *rq = tio->orig;
-
 	rq->cmd_flags |= REQ_FAILED;
-	dm_complete_request(clone, error);
+	dm_complete_request(rq, error);
 }
-EXPORT_SYMBOL_GPL(dm_kill_unmapped_request);
 
 /*
- * Called with the queue lock held
+ * Called with the clone's queue lock held
  */
 static void end_clone_request(struct request *clone, int error)
 {
-	/*
-	 * For just cleaning up the information of the queue in which
-	 * the clone was dispatched.
-	 * The clone is *NOT* freed actually here because it is alloced from
-	 * dm own mempool and REQ_ALLOCED isn't set in clone->cmd_flags.
-	 */
-	__blk_put_request(clone->q, clone);
+	struct dm_rq_target_io *tio = clone->end_io_data;
+
+	if (!clone->q->mq_ops) {
+		/*
+		 * For just cleaning up the information of the queue in which
+		 * the clone was dispatched.
+		 * The clone is *NOT* freed actually here because it is alloced
+		 * from dm own mempool (REQ_ALLOCED isn't set).
+		 */
+		__blk_put_request(clone->q, clone);
+	}
 
 	/*
 	 * Actual request completion is done in a softirq context which doesn't
-	 * hold the queue lock.  Otherwise, deadlock could occur because:
+	 * hold the clone's queue lock.  Otherwise, deadlock could occur because:
 	 *     - another request may be submitted by the upper level driver
 	 *       of the stacking during the completion
 	 *     - the submission which requires queue lock may be done
-	 *       against this queue
+	 *       against this clone's queue
 	 */
-	dm_complete_request(clone, error);
+	dm_complete_request(tio->orig, error);
 }
 
 /*
···
 		_dm_request(q, bio);
 }
 
-void dm_dispatch_request(struct request *rq)
+static void dm_dispatch_clone_request(struct request *clone, struct request *rq)
 {
 	int r;
 
-	if (blk_queue_io_stat(rq->q))
-		rq->cmd_flags |= REQ_IO_STAT;
+	if (blk_queue_io_stat(clone->q))
+		clone->cmd_flags |= REQ_IO_STAT;
 
-	rq->start_time = jiffies;
-	r = blk_insert_cloned_request(rq->q, rq);
+	clone->start_time = jiffies;
+	r = blk_insert_cloned_request(clone->q, clone);
 	if (r)
+		/* must complete clone in terms of original request */
 		dm_complete_request(rq, r);
 }
-EXPORT_SYMBOL_GPL(dm_dispatch_request);
 
 static int dm_rq_bio_constructor(struct bio *bio, struct bio *bio_orig,
 				 void *data)
···
 }
 
 static int setup_clone(struct request *clone, struct request *rq,
-		       struct dm_rq_target_io *tio)
+		       struct dm_rq_target_io *tio, gfp_t gfp_mask)
 {
 	int r;
 
-	blk_rq_init(NULL, clone);
-	r = blk_rq_prep_clone(clone, rq, tio->md->bs, GFP_ATOMIC,
+	r = blk_rq_prep_clone(clone, rq, tio->md->bs, gfp_mask,
 			      dm_rq_bio_constructor, tio);
 	if (r)
 		return r;
···
 	clone->end_io = end_clone_request;
 	clone->end_io_data = tio;
 
+	tio->clone = clone;
+
 	return 0;
 }
 
 static struct request *clone_rq(struct request *rq, struct mapped_device *md,
-				gfp_t gfp_mask)
+				struct dm_rq_target_io *tio, gfp_t gfp_mask)
 {
-	struct request *clone;
+	struct request *clone = alloc_clone_request(md, gfp_mask);
+
+	if (!clone)
+		return NULL;
+
+	blk_rq_init(NULL, clone);
+	if (setup_clone(clone, rq, tio, gfp_mask)) {
+		/* -ENOMEM */
+		free_clone_request(md, clone);
+		return NULL;
+	}
+
+	return clone;
+}
+
+static void map_tio_request(struct kthread_work *work);
+
+static struct dm_rq_target_io *prep_tio(struct request *rq,
+					struct mapped_device *md, gfp_t gfp_mask)
+{
 	struct dm_rq_target_io *tio;
+	int srcu_idx;
+	struct dm_table *table;
 
 	tio = alloc_rq_tio(md, gfp_mask);
 	if (!tio)
···
 
 	tio->md = md;
 	tio->ti = NULL;
+	tio->clone = NULL;
 	tio->orig = rq;
 	tio->error = 0;
 	memset(&tio->info, 0, sizeof(tio->info));
+	init_kthread_work(&tio->work, map_tio_request);
 
-	clone = &tio->clone;
-	if (setup_clone(clone, rq, tio)) {
-		/* -ENOMEM */
-		free_rq_tio(tio);
-		return NULL;
+	table = dm_get_live_table(md, &srcu_idx);
+	if (!dm_table_mq_request_based(table)) {
+		if (!clone_rq(rq, md, tio, gfp_mask)) {
+			dm_put_live_table(md, srcu_idx);
+			free_rq_tio(tio);
+			return NULL;
+		}
 	}
+	dm_put_live_table(md, srcu_idx);
 
-	return clone;
+	return tio;
 }
 
 /*
···
 static int dm_prep_fn(struct request_queue *q, struct request *rq)
 {
 	struct mapped_device *md = q->queuedata;
-	struct request *clone;
+	struct dm_rq_target_io *tio;
 
 	if (unlikely(rq->special)) {
 		DMWARN("Already has something in rq->special.");
 		return BLKPREP_KILL;
 	}
 
-	clone = clone_rq(rq, md, GFP_ATOMIC);
-	if (!clone)
+	tio = prep_tio(rq, md, GFP_ATOMIC);
+	if (!tio)
 		return BLKPREP_DEFER;
 
-	rq->special = clone;
+	rq->special = tio;
 	rq->cmd_flags |= REQ_DONTPREP;
 
 	return BLKPREP_OK;
···
 
 /*
  * Returns:
- * 0  : the request has been processed (not requeued)
- * !0 : the request has been requeued
+ * 0                : the request has been processed
+ * DM_MAPIO_REQUEUE : the original request needs to be requeued
+ * < 0              : the request was completed due to failure
  */
-static int map_request(struct dm_target *ti, struct request *clone,
+static int map_request(struct dm_target *ti, struct request *rq,
 		       struct mapped_device *md)
 {
-	int r, requeued = 0;
-	struct dm_rq_target_io *tio = clone->end_io_data;
+	int r;
+	struct dm_rq_target_io *tio = rq->special;
+	struct request *clone = NULL;
 
-	tio->ti = ti;
-	r = ti->type->map_rq(ti, clone, &tio->info);
+	if (tio->clone) {
+		clone = tio->clone;
+		r = ti->type->map_rq(ti, clone, &tio->info);
+	} else {
+		r = ti->type->clone_and_map_rq(ti, rq, &tio->info, &clone);
+		if (r < 0) {
+			/* The target wants to complete the I/O */
+			dm_kill_unmapped_request(rq, r);
+			return r;
+		}
+		if (IS_ERR(clone))
+			return DM_MAPIO_REQUEUE;
+		if (setup_clone(clone, rq, tio, GFP_KERNEL)) {
+			/* -ENOMEM */
+			ti->type->release_clone_rq(clone);
+			return DM_MAPIO_REQUEUE;
+		}
+	}
+
 	switch (r) {
 	case DM_MAPIO_SUBMITTED:
 		/* The target has taken the I/O to submit by itself later */
···
 	case DM_MAPIO_REMAPPED:
 		/* The target has remapped the I/O so dispatch it */
 		trace_block_rq_remap(clone->q, clone, disk_devt(dm_disk(md)),
-				     blk_rq_pos(tio->orig));
-		dm_dispatch_request(clone);
+				     blk_rq_pos(rq));
+		dm_dispatch_clone_request(clone, rq);
 		break;
 	case DM_MAPIO_REQUEUE:
 		/* The target wants to requeue the I/O */
 		dm_requeue_unmapped_request(clone);
-		requeued = 1;
 		break;
 	default:
 		if (r > 0) {
···
 		}
 
 		/* The target wants to complete the I/O */
-		dm_kill_unmapped_request(clone, r);
-		break;
+		dm_kill_unmapped_request(rq, r);
+		return r;
 	}
 
-	return requeued;
+	return 0;
 }
 
-static struct request *dm_start_request(struct mapped_device *md, struct request *orig)
+static void map_tio_request(struct kthread_work *work)
 {
-	struct request *clone;
+	struct dm_rq_target_io *tio = container_of(work, struct dm_rq_target_io, work);
+	struct request *rq = tio->orig;
+	struct mapped_device *md = tio->md;
 
+	if (map_request(tio->ti, rq, md) == DM_MAPIO_REQUEUE)
+		dm_requeue_unmapped_original_request(md, rq);
+}
+
+static void dm_start_request(struct mapped_device *md, struct request *orig)
+{
 	blk_start_request(orig);
-	clone = orig->special;
-	atomic_inc(&md->pending[rq_data_dir(clone)]);
+	atomic_inc(&md->pending[rq_data_dir(orig)]);
 
 	/*
 	 * Hold the md reference here for the in-flight I/O.
···
 	 * See the comment in rq_completed() too.
 	 */
 	dm_get(md);
-
-	return clone;
 }
 
 /*
···
 	int srcu_idx;
 	struct dm_table *map = dm_get_live_table(md, &srcu_idx);
 	struct dm_target *ti;
-	struct request *rq, *clone;
+	struct request *rq;
+	struct dm_rq_target_io *tio;
 	sector_t pos;
 
 	/*
···
 		ti = dm_table_find_target(map, pos);
 		if (!dm_target_is_valid(ti)) {
 			/*
-			 * Must perform setup, that dm_done() requires,
+			 * Must perform setup, that rq_completed() requires,
 			 * before calling dm_kill_unmapped_request
 			 */
 			DMERR_LIMIT("request attempted access beyond the end of device");
-			clone = dm_start_request(md, rq);
-			dm_kill_unmapped_request(clone, -EIO);
+			dm_start_request(md, rq);
+			dm_kill_unmapped_request(rq, -EIO);
 			continue;
 		}
 
 		if (ti->type->busy && ti->type->busy(ti))
 			goto delay_and_out;
 
-		clone = dm_start_request(md, rq);
+		dm_start_request(md, rq);
 
-		spin_unlock(q->queue_lock);
-		if (map_request(ti, clone, md))
-			goto requeued;
-
+		tio = rq->special;
+		/* Establish tio->ti before queuing work (map_tio_request) */
+		tio->ti = ti;
+		queue_kthread_work(&md->kworker, &tio->work);
 		BUG_ON(!irqs_disabled());
-		spin_lock(q->queue_lock);
 	}
 
 	goto out;
-
-requeued:
-	BUG_ON(!irqs_disabled());
-	spin_lock(q->queue_lock);
 
 delay_and_out:
 	blk_delay_queue(q, HZ / 10);
···
 	INIT_WORK(&md->work, dm_wq_work);
 	init_waitqueue_head(&md->eventq);
 	init_completion(&md->kobj_holder.completion);
+	md->kworker_task = NULL;
 
 	md->disk->major = _major;
 	md->disk->first_minor = minor;
···
 	unlock_fs(md);
 	bdput(md->bdev);
 	destroy_workqueue(md->wq);
+
+	if (md->kworker_task)
+		kthread_stop(md->kworker_task);
 	if (md->io_pool)
 		mempool_destroy(md->io_pool);
+	if (md->rq_pool)
+		mempool_destroy(md->rq_pool);
 	if (md->bs)
 		bioset_free(md->bs);
 	blk_integrity_unregister(md->disk);
···
 			bioset_free(md->bs);
 			md->bs = p->bs;
 			p->bs = NULL;
-		} else if (dm_table_get_type(t) == DM_TYPE_REQUEST_BASED) {
-			/*
-			 * There's no need to reload with request-based dm
-			 * because the size of front_pad doesn't change.
-			 * Note for future: If you are to reload bioset,
-			 * prep-ed requests in the queue may refer
-			 * to bio from the old bioset, so you must walk
-			 * through the queue to unprep.
-			 */
 		}
+		/*
+		 * There's no need to reload with request-based dm
+		 * because the size of front_pad doesn't change.
+		 * Note for future: If you are to reload bioset,
+		 * prep-ed requests in the queue may refer
+		 * to bio from the old bioset, so you must walk
+		 * through the queue to unprep.
+		 */
 		goto out;
 	}
 
-	BUG_ON(!p || md->io_pool || md->bs);
+	BUG_ON(!p || md->io_pool || md->rq_pool || md->bs);
 
 	md->io_pool = p->io_pool;
 	p->io_pool = NULL;
+	md->rq_pool = p->rq_pool;
+	p->rq_pool = NULL;
 	md->bs = p->bs;
 	p->bs = NULL;
 
···
 	return md->type;
 }
 
+static bool dm_md_type_request_based(struct mapped_device *md)
+{
+	unsigned table_type = dm_get_md_type(md);
+
+	return (table_type == DM_TYPE_REQUEST_BASED ||
+		table_type == DM_TYPE_MQ_REQUEST_BASED);
+}
+
 struct target_type *dm_get_immutable_target_type(struct mapped_device *md)
 {
 	return md->immutable_target_type;
···
 	blk_queue_prep_rq(md->queue, dm_prep_fn);
 	blk_queue_lld_busy(md->queue, dm_lld_busy);
 
+	/* Also initialize the request-based DM worker thread */
+	init_kthread_worker(&md->kworker);
+	md->kworker_task = kthread_run(kthread_worker_fn, &md->kworker,
+				       "kdmwork-%s", dm_device_name(md));
+
 	elv_register_queue(md->queue);
 
 	return 1;
···
  */
 int dm_setup_md_queue(struct mapped_device *md)
 {
-	if ((dm_get_md_type(md) == DM_TYPE_REQUEST_BASED) &&
-	    !dm_init_request_based_queue(md)) {
+	if (dm_md_type_request_based(md) && !dm_init_request_based_queue(md)) {
 		DMWARN("Cannot initialize queue for request-based mapped device");
 		return -EINVAL;
 	}
···
 	idr_replace(&_minor_idr, MINOR_ALLOCED, MINOR(disk_devt(dm_disk(md))));
 	set_bit(DMF_FREEING, &md->flags);
 	spin_unlock(&_minor_lock);
+
+	if (dm_request_based(md))
+		flush_kthread_worker(&md->kworker);
 
 	if (!dm_suspended_md(md)) {
 		dm_table_presuspend_targets(map);
···
 	 * Stop md->queue before flushing md->wq in case request-based
 	 * dm defers requests to md->wq from md->queue.
 	 */
-	if (dm_request_based(md))
+	if (dm_request_based(md)) {
 		stop_queue(md->queue);
+		flush_kthread_worker(&md->kworker);
+	}
 
 	flush_workqueue(md->wq);
 
···
 {
 	struct dm_md_mempools *pools = kzalloc(sizeof(*pools), GFP_KERNEL);
 	struct kmem_cache *cachep;
-	unsigned int pool_size;
+	unsigned int pool_size = 0;
 	unsigned int front_pad;
 
 	if (!pools)
 		return NULL;
 
-	if (type == DM_TYPE_BIO_BASED) {
+	switch (type) {
+	case DM_TYPE_BIO_BASED:
 		cachep = _io_cache;
 		pool_size = dm_get_reserved_bio_based_ios();
 		front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone);
-	} else if (type == DM_TYPE_REQUEST_BASED) {
-		cachep = _rq_tio_cache;
+		break;
+	case DM_TYPE_REQUEST_BASED:
 		pool_size = dm_get_reserved_rq_based_ios();
+		pools->rq_pool = mempool_create_slab_pool(pool_size, _rq_cache);
+		if (!pools->rq_pool)
+			goto out;
+		/* fall through to setup remaining rq-based pools */
+	case DM_TYPE_MQ_REQUEST_BASED:
+		cachep = _rq_tio_cache;
+		if (!pool_size)
+			pool_size = dm_get_reserved_rq_based_ios();
 		front_pad = offsetof(struct dm_rq_clone_bio_info, clone);
 		/* per_bio_data_size is not used. See __bind_mempools(). */
 		WARN_ON(per_bio_data_size != 0);
-	} else
+		break;
+	default:
 		goto out;
+	}
 
 	pools->io_pool = mempool_create_slab_pool(pool_size, cachep);
 	if (!pools->io_pool)
···
 
 	if (pools->io_pool)
 		mempool_destroy(pools->io_pool);
+
+	if (pools->rq_pool)
+		mempool_destroy(pools->rq_pool);
 
 	if (pools->bs)
 		bioset_free(pools->bs);

drivers/md/dm.h

···
 /*
  * Type of table and mapped_device's mempool
  */
-#define DM_TYPE_NONE		0
-#define DM_TYPE_BIO_BASED	1
-#define DM_TYPE_REQUEST_BASED	2
+#define DM_TYPE_NONE			0
+#define DM_TYPE_BIO_BASED		1
+#define DM_TYPE_REQUEST_BASED		2
+#define DM_TYPE_MQ_REQUEST_BASED	3
 
 /*
  * List of devices that a metadevice uses and should open/close.
···
 unsigned dm_table_get_type(struct dm_table *t);
 struct target_type *dm_table_get_immutable_target_type(struct dm_table *t);
 bool dm_table_request_based(struct dm_table *t);
+bool dm_table_mq_request_based(struct dm_table *t);
 void dm_table_free_md_mempools(struct dm_table *t);
 struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t);
 
···
 /*
  * To check whether the target type is request-based or not (bio-based).
  */
-#define dm_target_request_based(t) ((t)->type->map_rq != NULL)
+#define dm_target_request_based(t) (((t)->type->map_rq != NULL) || \
+				    ((t)->type->clone_and_map_rq != NULL))
 
 /*
  * To check whether the target type is a hybrid (capable of being

include/linux/device-mapper.h

···
 typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio);
 typedef int (*dm_map_request_fn) (struct dm_target *ti, struct request *clone,
 				  union map_info *map_context);
+typedef int (*dm_clone_and_map_request_fn) (struct dm_target *ti,
+					    struct request *rq,
+					    union map_info *map_context,
+					    struct request **clone);
+typedef void (*dm_release_clone_request_fn) (struct request *clone);
 
 /*
  * Returns:
···
 	dm_dtr_fn dtr;
 	dm_map_fn map;
 	dm_map_request_fn map_rq;
+	dm_clone_and_map_request_fn clone_and_map_rq;
+	dm_release_clone_request_fn release_clone_rq;
 	dm_endio_fn end_io;
 	dm_request_endio_fn rq_end_io;
 	dm_presuspend_fn presuspend;
···
 /*-----------------------------------------------------------------
  * Helper for block layer and dm core operations
  *---------------------------------------------------------------*/
-void dm_dispatch_request(struct request *rq);
-void dm_requeue_unmapped_request(struct request *rq);
-void dm_kill_unmapped_request(struct request *rq, int error);
 int dm_underlying_device_busy(struct request_queue *q);
 
 #endif	/* _LINUX_DEVICE_MAPPER_H */

include/uapi/linux/dm-ioctl.h

···
 #define DM_DEV_SET_GEOMETRY	_IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl)
 
 #define DM_VERSION_MAJOR	4
-#define DM_VERSION_MINOR	29
+#define DM_VERSION_MINOR	30
 #define DM_VERSION_PATCHLEVEL	0
-#define DM_VERSION_EXTRA	"-ioctl (2014-10-28)"
+#define DM_VERSION_EXTRA	"-ioctl (2014-12-22)"
 
 /* Status bits */
 #define DM_READONLY_FLAG	(1 << 0) /* In/Out */