dm: remove support for stacking dm-mq on .request_fn device(s)

Remove all fiddley code that propped up this support for a blk-mq
request-queue ontop of all .request_fn devices.

Testing has proven this niche request-based dm-mq mode to be buggy, when
testing fault tolerance with DM multipath, and there is no point trying
to preserve it.

Should help improve efficiency of pure dm-mq code and make code
maintenance less delicate.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>

2 files changed, 20 insertions, 40 deletions

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 3ddaa11584ef..7259eeba6a58 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -418,7 +418,10 @@ static int __multipath_map(struct dm_target *ti, struct request *clone,
         spin_unlock_irq(&m->lock);
 
         if (clone) {
-                /* Old request-based interface: allocated clone is passed in */
+                /*
+                 * Old request-based interface: allocated clone is passed in.
+                 * Used by: .request_fn stacked on .request_fn path(s).
+                 */
                 clone->q = bdev_get_queue(bdev);
                 clone->rq_disk = bdev->bd_disk;
                 clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 8a62e43fbff5..26fedd93702e 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1141,11 +1141,6 @@ static void free_rq_clone(struct request *clone)
         else if (!md->queue->mq_ops)
                 /* request_fn queue stacked on request_fn queue(s) */
                 free_clone_request(md, clone);
-        /*
-         * NOTE: for the blk-mq queue stacked on request_fn queue(s) case:
-         * no need to call free_clone_request() because we leverage blk-mq by
-         * allocating the clone at the end of the blk-mq pdu (see: clone_rq)
-         */
 
         if (!md->queue->mq_ops)
                 free_rq_tio(tio);
@@ -1866,24 +1861,18 @@ static struct request *clone_rq(struct request *rq, struct mapped_device *md,
                                 struct dm_rq_target_io *tio, gfp_t gfp_mask)
 {
         /*
-         * Do not allocate a clone if tio->clone was already set
-         * (see: dm_mq_queue_rq).
+         * Create clone for use with .request_fn request_queue
          */
-        bool alloc_clone = !tio->clone;
         struct request *clone;
 
-        if (alloc_clone) {
-                clone = alloc_clone_request(md, gfp_mask);
-                if (!clone)
-                        return NULL;
-        } else
-                clone = tio->clone;
+        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 */
-                if (alloc_clone)
-                        free_clone_request(md, clone);
+                free_clone_request(md, clone);
                 return NULL;
         }
 
@@ -2692,22 +2681,12 @@ static int dm_mq_queue_rq(struct blk_mq_hw_ctx *hctx,
          */
         tio->ti = ti;
 
-        /*
-         * Both the table and md type cannot change after initial table load
-         */
-        if (dm_get_md_type(md) == DM_TYPE_REQUEST_BASED) {
-                /* clone request is allocated at the end of the pdu */
-                tio->clone = (void *)blk_mq_rq_to_pdu(rq) + sizeof(struct dm_rq_target_io);
-                (void) clone_rq(rq, md, tio, GFP_ATOMIC);
-                queue_kthread_work(&md->kworker, &tio->work);
-        } else {
-                /* Direct call is fine since .queue_rq allows allocations */
-                if (map_request(tio, rq, md) == DM_MAPIO_REQUEUE) {
-                        /* Undo dm_start_request() before requeuing */
-                        rq_end_stats(md, rq);
-                        rq_completed(md, rq_data_dir(rq), false);
-                        return BLK_MQ_RQ_QUEUE_BUSY;
-                }
+        /* Direct call is fine since .queue_rq allows allocations */
+        if (map_request(tio, rq, md) == DM_MAPIO_REQUEUE) {
+                /* Undo dm_start_request() before requeuing */
+                rq_end_stats(md, rq);
+                rq_completed(md, rq_data_dir(rq), false);
+                return BLK_MQ_RQ_QUEUE_BUSY;
         }
 
         return BLK_MQ_RQ_QUEUE_OK;
@@ -2726,6 +2705,11 @@ static int dm_init_request_based_blk_mq_queue(struct mapped_device *md)
         struct request_queue *q;
         int err;
 
+        if (dm_get_md_type(md) == DM_TYPE_REQUEST_BASED) {
+                DMERR("request-based dm-mq may only be stacked on blk-mq device(s)");
+                return -EINVAL;
+        }
+
         md->tag_set = kzalloc(sizeof(struct blk_mq_tag_set), GFP_KERNEL);
         if (!md->tag_set)
                 return -ENOMEM;
@@ -2738,10 +2722,6 @@ static int dm_init_request_based_blk_mq_queue(struct mapped_device *md)
         md->tag_set->driver_data = md;
 
         md->tag_set->cmd_size = sizeof(struct dm_rq_target_io);
-        if (md_type == DM_TYPE_REQUEST_BASED) {
-                /* put the memory for non-blk-mq clone at the end of the pdu */
-                md->tag_set->cmd_size += sizeof(struct request);
-        }
 
         err = blk_mq_alloc_tag_set(md->tag_set);
         if (err)
@@ -2758,9 +2738,6 @@ static int dm_init_request_based_blk_mq_queue(struct mapped_device *md)
         /* backfill 'mq' sysfs registration normally done in blk_register_queue */
         blk_mq_register_disk(md->disk);
 
-        if (md_type == DM_TYPE_REQUEST_BASED)
-                init_rq_based_worker_thread(md);
-
         return 0;
 
 out_tag_set: