dm: allocate requests in target when stacking on blk-mq devices

For blk-mq request-based DM the responsibility of allocating a cloned
request is transfered from DM core to the target type. Doing so
enables the cloned request to be allocated from the appropriate
blk-mq request_queue's pool (only the DM target, e.g. multipath, can
know which block device to send a given cloned request to).

Care was taken to preserve compatibility with old-style block request
completion that requires request-based DM _not_ acquire the clone
request's queue lock in the completion path. As such, there are now 2
different request-based DM target_type interfaces:
1) the original .map_rq() interface will continue to be used for
non-blk-mq devices -- the preallocated clone request is passed in
from DM core.
2) a new .clone_and_map_rq() and .release_clone_rq() will be used for
blk-mq devices -- blk_get_request() and blk_put_request() are used
respectively from these hooks.

dm_table_set_type() was updated to detect if the request-based target is
being stacked on blk-mq devices, if so DM_TYPE_MQ_REQUEST_BASED is set.
DM core disallows switching the DM table's type after it is set. This
means that there is no mixing of non-blk-mq and blk-mq devices within
the same request-based DM table.

[This patch was started by Keith and later heavily modified by Mike]

Tested-by: Bart Van Assche <[e-mail]>
Signed-off-by: Keith Busch <[e-mail]>
Signed-off-by: Mike Snitzer <[e-mail]>

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);
+	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;
···
 
 	bdev = pgpath->path.dev->bdev;
 
-	clone->q = bdev_get_queue(bdev);
-	clone->rq_disk = bdev->bd_disk;
-	clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
-
 	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,
···
 	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,

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;
···
 	/* 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");
+		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;
+			}
 	}
 
 	/*
···
 		return -EINVAL;
 	}
 
-	t->type = DM_TYPE_REQUEST_BASED;
+	t->type = !use_blk_mq ? DM_TYPE_REQUEST_BASED : DM_TYPE_MQ_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)

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.c

···
 	struct dm_rq_target_io *tio = clone->end_io_data;
 
 	blk_rq_unprep_clone(clone);
-	free_clone_request(tio->md, 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);
 }
 
···
 	rq->special = NULL;
 	rq->cmd_flags &= ~REQ_DONTPREP;
 
-	free_rq_clone(clone);
+	if (clone)
+		free_rq_clone(clone);
 }
 
 /*
···
 	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 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.
  */
 static void dm_kill_unmapped_request(struct request *rq, int error)
 {
···
 {
 	struct dm_rq_target_io *tio = clone->end_io_data;
 
-	/*
-	 * 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);
+	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
···
 					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)
···
 	memset(&tio->info, 0, sizeof(tio->info));
 	init_kthread_work(&tio->work, map_tio_request);
 
-	if (!clone_rq(rq, md, tio, gfp_mask)) {
-		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 tio;
 }
···
 
 /*
  * 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 *rq,
 		       struct mapped_device *md)
 {
-	int r, requeued = 0;
+	int r;
 	struct dm_rq_target_io *tio = rq->special;
-	struct request *clone = tio->clone;
+	struct request *clone = NULL;
 
-	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_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(rq, r);
-		break;
+		return r;
 	}
 
-	return requeued;
+	return 0;
 }
 
 static void map_tio_request(struct kthread_work *work)
 {
 	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;
 
-	map_request(tio->ti, tio->orig, 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)
···
 	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;
···
  */
 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;
 	}
···
 {
 	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)

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);
 

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;

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 */