diff --git a/block/bio.c b/block/bio.c
index 8c2e55e39a1b9a6eca0c169f658b35a8006f5372..0ec61c9e536c2032778db165c7b31d98bdc5e735 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -746,6 +746,14 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 
 			goto done;
 		}
+
+		/*
+		 * If the queue doesn't support SG gaps and adding this
+		 * offset would create a gap, disallow it.
+		 */
+		if (q->queue_flags & (1 << QUEUE_FLAG_SG_GAPS) &&
+		    bvec_gap_to_prev(prev, offset))
+			return 0;
 	}
 
 	if (bio->bi_vcnt >= bio->bi_max_vecs)
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 069bc202ffe340fb69e9e3dc9414fe8b03043bd1..b9f4cc494ecefbf2560483bdc4685bb98d59b5d3 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -80,7 +80,7 @@ static struct blkcg_gq *blkg_alloc(struct blkcg *blkcg, struct request_queue *q,
 	blkg->q = q;
 	INIT_LIST_HEAD(&blkg->q_node);
 	blkg->blkcg = blkcg;
-	blkg->refcnt = 1;
+	atomic_set(&blkg->refcnt, 1);
 
 	/* root blkg uses @q->root_rl, init rl only for !root blkgs */
 	if (blkcg != &blkcg_root) {
@@ -399,11 +399,8 @@ void __blkg_release_rcu(struct rcu_head *rcu_head)
 
 	/* release the blkcg and parent blkg refs this blkg has been holding */
 	css_put(&blkg->blkcg->css);
-	if (blkg->parent) {
-		spin_lock_irq(blkg->q->queue_lock);
+	if (blkg->parent)
 		blkg_put(blkg->parent);
-		spin_unlock_irq(blkg->q->queue_lock);
-	}
 
 	blkg_free(blkg);
 }
@@ -1093,7 +1090,7 @@ EXPORT_SYMBOL_GPL(blkcg_deactivate_policy);
  * Register @pol with blkcg core.  Might sleep and @pol may be modified on
  * successful registration.  Returns 0 on success and -errno on failure.
  */
-int __init blkcg_policy_register(struct blkcg_policy *pol)
+int blkcg_policy_register(struct blkcg_policy *pol)
 {
 	int i, ret;
 
diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
index cbb7f943f78a170c12d97429ce582562e97b4485..d3fd7aa3d2a369f9ef0017ece713965c95529e82 100644
--- a/block/blk-cgroup.h
+++ b/block/blk-cgroup.h
@@ -18,6 +18,7 @@
 #include <linux/seq_file.h>
 #include <linux/radix-tree.h>
 #include <linux/blkdev.h>
+#include <linux/atomic.h>
 
 /* Max limits for throttle policy */
 #define THROTL_IOPS_MAX		UINT_MAX
@@ -104,7 +105,7 @@ struct blkcg_gq {
 	struct request_list		rl;
 
 	/* reference count */
-	int				refcnt;
+	atomic_t			refcnt;
 
 	/* is this blkg online? protected by both blkcg and q locks */
 	bool				online;
@@ -145,7 +146,7 @@ void blkcg_drain_queue(struct request_queue *q);
 void blkcg_exit_queue(struct request_queue *q);
 
 /* Blkio controller policy registration */
-int __init blkcg_policy_register(struct blkcg_policy *pol);
+int blkcg_policy_register(struct blkcg_policy *pol);
 void blkcg_policy_unregister(struct blkcg_policy *pol);
 int blkcg_activate_policy(struct request_queue *q,
 			  const struct blkcg_policy *pol);
@@ -257,13 +258,12 @@ static inline int blkg_path(struct blkcg_gq *blkg, char *buf, int buflen)
  * blkg_get - get a blkg reference
  * @blkg: blkg to get
  *
- * The caller should be holding queue_lock and an existing reference.
+ * The caller should be holding an existing reference.
  */
 static inline void blkg_get(struct blkcg_gq *blkg)
 {
-	lockdep_assert_held(blkg->q->queue_lock);
-	WARN_ON_ONCE(!blkg->refcnt);
-	blkg->refcnt++;
+	WARN_ON_ONCE(atomic_read(&blkg->refcnt) <= 0);
+	atomic_inc(&blkg->refcnt);
 }
 
 void __blkg_release_rcu(struct rcu_head *rcu);
@@ -271,14 +271,11 @@ void __blkg_release_rcu(struct rcu_head *rcu);
 /**
  * blkg_put - put a blkg reference
  * @blkg: blkg to put
- *
- * The caller should be holding queue_lock.
  */
 static inline void blkg_put(struct blkcg_gq *blkg)
 {
-	lockdep_assert_held(blkg->q->queue_lock);
-	WARN_ON_ONCE(blkg->refcnt <= 0);
-	if (!--blkg->refcnt)
+	WARN_ON_ONCE(atomic_read(&blkg->refcnt) <= 0);
+	if (atomic_dec_and_test(&blkg->refcnt))
 		call_rcu(&blkg->rcu_head, __blkg_release_rcu);
 }
 
@@ -580,7 +577,7 @@ static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, void *key) { ret
 static inline int blkcg_init_queue(struct request_queue *q) { return 0; }
 static inline void blkcg_drain_queue(struct request_queue *q) { }
 static inline void blkcg_exit_queue(struct request_queue *q) { }
-static inline int __init blkcg_policy_register(struct blkcg_policy *pol) { return 0; }
+static inline int blkcg_policy_register(struct blkcg_policy *pol) { return 0; }
 static inline void blkcg_policy_unregister(struct blkcg_policy *pol) { }
 static inline int blkcg_activate_policy(struct request_queue *q,
 					const struct blkcg_policy *pol) { return 0; }
diff --git a/block/blk-merge.c b/block/blk-merge.c
index b3bf0df0f4c2743aa99ef7ac179db324eb5e4657..54535831f1e197a6a6c56a5cc8a4b08e90d4020b 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -568,6 +568,8 @@ int blk_attempt_req_merge(struct request_queue *q, struct request *rq,
 
 bool blk_rq_merge_ok(struct request *rq, struct bio *bio)
 {
+	struct request_queue *q = rq->q;
+
 	if (!rq_mergeable(rq) || !bio_mergeable(bio))
 		return false;
 
@@ -591,6 +593,14 @@ bool blk_rq_merge_ok(struct request *rq, struct bio *bio)
 	    !blk_write_same_mergeable(rq->bio, bio))
 		return false;
 
+	if (q->queue_flags & (1 << QUEUE_FLAG_SG_GAPS)) {
+		struct bio_vec *bprev;
+
+		bprev = &rq->biotail->bi_io_vec[bio->bi_vcnt - 1];
+		if (bvec_gap_to_prev(bprev, bio->bi_io_vec[0].bv_offset))
+			return false;
+	}
+
 	return true;
 }
 
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 0ef2dc7f01bf725884e9fbdff2dd5ea2a852f15b..ad69ef657e850cc79c6379667c0c501f9400a551 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -878,7 +878,7 @@ void blk_mq_start_hw_queue(struct blk_mq_hw_ctx *hctx)
 	clear_bit(BLK_MQ_S_STOPPED, &hctx->state);
 
 	preempt_disable();
-	__blk_mq_run_hw_queue(hctx);
+	blk_mq_run_hw_queue(hctx, false);
 	preempt_enable();
 }
 EXPORT_SYMBOL(blk_mq_start_hw_queue);
diff --git a/block/elevator.c b/block/elevator.c
index 34bded18910e74f73754b9584361deba40611682..24c28b659bb34f00eae0e4b900266e1dca069169 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -825,7 +825,7 @@ void elv_unregister_queue(struct request_queue *q)
 }
 EXPORT_SYMBOL(elv_unregister_queue);
 
-int __init elv_register(struct elevator_type *e)
+int elv_register(struct elevator_type *e)
 {
 	char *def = "";
 
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index b6c8aaf4931bc8434635e74efa004f9ddf8c0304..5b17ec88ea058e766071e66eeadf3d8fca3f4940 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -1337,8 +1337,11 @@ int drbd_submit_peer_request(struct drbd_device *device,
 		return 0;
 	}
 
+	/* Discards don't have any payload.
+	 * But the scsi layer still expects a bio_vec it can use internally,
+	 * see sd_setup_discard_cmnd() and blk_add_request_payload(). */
 	if (peer_req->flags & EE_IS_TRIM)
-		nr_pages = 0; /* discards don't have any payload. */
+		nr_pages = 1;
 
 	/* In most cases, we will only need one bio.  But in case the lower
 	 * level restrictions happen to be different at this offset on this
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 677db049f55a13fa692ba8a1e4bbe7ea092c3dc1..56d46ffb08e1512eba11e2b1628baeb3a420d63e 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -3777,7 +3777,7 @@ static void floppy_rb0_cb(struct bio *bio, int err)
 	int drive = cbdata->drive;
 
 	if (err) {
-		pr_info("floppy: error %d while reading block 0", err);
+		pr_info("floppy: error %d while reading block 0\n", err);
 		set_bit(FD_OPEN_SHOULD_FAIL_BIT, &UDRS->flags);
 	}
 	complete(&cbdata->complete);
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 5a645769f020f956415c1fac11ac868a1922f161..d2633ee099d975836b1fc457cd58e323bfcc8784 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -186,6 +186,15 @@ static inline void *bio_data(struct bio *bio)
 #define BIOVEC_SEG_BOUNDARY(q, b1, b2) \
 	__BIO_SEG_BOUNDARY(bvec_to_phys((b1)), bvec_to_phys((b2)) + (b2)->bv_len, queue_segment_boundary((q)))
 
+/*
+ * Check if adding a bio_vec after bprv with offset would create a gap in
+ * the SG list. Most drivers don't care about this, but some do.
+ */
+static inline bool bvec_gap_to_prev(struct bio_vec *bprv, unsigned int offset)
+{
+	return offset || ((bprv->bv_offset + bprv->bv_len) & (PAGE_SIZE - 1));
+}
+
 #define bio_io_error(bio) bio_endio((bio), -EIO)
 
 /*
@@ -644,10 +653,6 @@ struct biovec_slab {
 
 #if defined(CONFIG_BLK_DEV_INTEGRITY)
 
-
-
-#define bip_vec_idx(bip, idx)	(&(bip->bip_vec[(idx)]))
-
 #define bip_for_each_vec(bvl, bip, iter)				\
 	for_each_bvec(bvl, (bip)->bip_vec, iter, (bip)->bip_iter)
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 713f8b62b435ea2cd2d273fc02c2b39520886d7d..8699bcf5f0999db98a8f2a2917c284d950a12d75 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -512,6 +512,7 @@ struct request_queue {
 #define QUEUE_FLAG_DEAD        19	/* queue tear-down finished */
 #define QUEUE_FLAG_INIT_DONE   20	/* queue is initialized */
 #define QUEUE_FLAG_NO_SG_MERGE 21	/* don't attempt to merge SG segments*/
+#define QUEUE_FLAG_SG_GAPS     22	/* queue doesn't support SG gaps */
 
 #define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
 				 (1 << QUEUE_FLAG_STACKABLE)	|	\
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index e2a6bd7fb13380bfa4b0d14251b8a64908f8b0f1..45a91474487daff11a69bf55a31d7cd146c75f72 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -143,7 +143,7 @@ extern void elv_drain_elevator(struct request_queue *);
  * io scheduler registration
  */
 extern void __init load_default_elevator_module(void);
-extern int __init elv_register(struct elevator_type *);
+extern int elv_register(struct elevator_type *);
 extern void elv_unregister(struct elevator_type *);
 
 /*