Merge branch 'for-4.15/block' of git://git.kernel.dk/linux-block

Pull core block layer updates from Jens Axboe: "This is the main pull request for block storage for 4.15-rc1. Nothing out of the ordinary in here, and no API changes or anything like that. Just various new features for drivers, core changes, etc. In particular, this pull request contains: - A patch series from Bart, closing the whole on blk/scsi-mq queue quescing. - A series from Christoph, building towards hidden gendisks (for multipath) and ability to move bio chains around. - NVMe - Support for native multipath for NVMe (Christoph). - Userspace notifications for AENs (Keith). - Command side-effects support (Keith). - SGL support (Chaitanya Kulkarni) - FC fixes and improvements (James Smart) - Lots of fixes and tweaks (Various) - bcache - New maintainer (Michael Lyle) - Writeback control improvements (Michael) - Various fixes (Coly, Elena, Eric, Liang, et al) - lightnvm updates, mostly centered around the pblk interface (Javier, Hans, and Rakesh). - Removal of unused bio/bvec kmap atomic interfaces (me, Christoph) - Writeback series that fix the much discussed hundreds of millions of sync-all units. This goes all the way, as discussed previously (me). - Fix for missing wakeup on writeback timer adjustments (Yafang Shao). - Fix laptop mode on blk-mq (me). - {mq,name} tupple lookup for IO schedulers, allowing us to have alias names. This means you can use 'deadline' on both !mq and on mq (where it's called mq-deadline). (me). - blktrace race fix, oopsing on sg load (me). - blk-mq optimizations (me). - Obscure waitqueue race fix for kyber (Omar). - NBD fixes (Josef). - Disable writeback throttling by default on bfq, like we do on cfq (Luca Miccio). - Series from Ming that enable us to treat flush requests on blk-mq like any other request. This is a really nice cleanup. - Series from Ming that improves merging on blk-mq with schedulers, getting us closer to flipping the switch on scsi-mq again. - BFQ updates (Paolo). - blk-mq atomic flags memory ordering fixes (Peter Z). - Loop cgroup support (Shaohua). - Lots of minor fixes from lots of different folks, both for core and driver code" * 'for-4.15/block' of git://git.kernel.dk/linux-block: (294 commits) nvme: fix visibility of "uuid" ns attribute blk-mq: fixup some comment typos and lengths ide: ide-atapi: fix compile error with defining macro DEBUG blk-mq: improve tag waiting setup for non-shared tags brd: remove unused brd_mutex blk-mq: only run the hardware queue if IO is pending block: avoid null pointer dereference on null disk fs: guard_bio_eod() needs to consider partitions xtensa/simdisk: fix compile error nvme: expose subsys attribute to sysfs nvme: create 'slaves' and 'holders' entries for hidden controllers block: create 'slaves' and 'holders' entries for hidden gendisks nvme: also expose the namespace identification sysfs files for mpath nodes nvme: implement multipath access to nvme subsystems nvme: track shared namespaces nvme: introduce a nvme_ns_ids structure nvme: track subsystems block, nvme: Introduce blk_mq_req_flags_t block, scsi: Make SCSI quiesce and resume work reliably block: Add the QUEUE_FLAG_PREEMPT_ONLY request queue flag ...
author: Linus Torvalds <torvalds@linux-foundation.org> 2017-11-14 15:32:19 -0800
committer: Linus Torvalds <torvalds@linux-foundation.org> 2017-11-14 15:32:19 -0800
commit: e2c5923c349c1738fe8fda980874d93f6fb2e5b6 (patch)
tree: b97a90170c45211bcc437761653aa8016c34afcd /block
parent: abc36be236358162202e86ad88616ff95a755101 (diff)
parent: a04b5de5050ab8b891128eb2c47a0916fe8622e1 (diff)
download: linux-e2c5923c349c1738fe8fda980874d93f6fb2e5b6.tar.bz2
26 files changed, 1097 insertions, 616 deletions
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index a4783da90ba8..889a8549d97f 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -108,6 +108,7 @@
 #include "blk-mq-tag.h"
 #include "blk-mq-sched.h"
 #include "bfq-iosched.h"
+#include "blk-wbt.h"
 
 #define BFQ_BFQQ_FNS(name)						\
 void bfq_mark_bfqq_##name(struct bfq_queue *bfqq)			\
@@ -724,6 +725,44 @@ static void bfq_updated_next_req(struct bfq_data *bfqd,
 	}
 }
 
+static unsigned int bfq_wr_duration(struct bfq_data *bfqd)
+{
+	u64 dur;
+
+	if (bfqd->bfq_wr_max_time > 0)
+		return bfqd->bfq_wr_max_time;
+
+	dur = bfqd->RT_prod;
+	do_div(dur, bfqd->peak_rate);
+
+	/*
+	 * Limit duration between 3 and 13 seconds. Tests show that
+	 * higher values than 13 seconds often yield the opposite of
+	 * the desired result, i.e., worsen responsiveness by letting
+	 * non-interactive and non-soft-real-time applications
+	 * preserve weight raising for a too long time interval.
+	 *
+	 * On the other end, lower values than 3 seconds make it
+	 * difficult for most interactive tasks to complete their jobs
+	 * before weight-raising finishes.
+	 */
+	if (dur > msecs_to_jiffies(13000))
+		dur = msecs_to_jiffies(13000);
+	else if (dur < msecs_to_jiffies(3000))
+		dur = msecs_to_jiffies(3000);
+
+	return dur;
+}
+
+/* switch back from soft real-time to interactive weight raising */
+static void switch_back_to_interactive_wr(struct bfq_queue *bfqq,
+					  struct bfq_data *bfqd)
+{
+	bfqq->wr_coeff = bfqd->bfq_wr_coeff;
+	bfqq->wr_cur_max_time = bfq_wr_duration(bfqd);
+	bfqq->last_wr_start_finish = bfqq->wr_start_at_switch_to_srt;
+}
+
 static void
 bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd,
 		      struct bfq_io_cq *bic, bool bfq_already_existing)
@@ -750,10 +789,16 @@ bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd,
 	if (bfqq->wr_coeff > 1 && (bfq_bfqq_in_large_burst(bfqq) ||
 	    time_is_before_jiffies(bfqq->last_wr_start_finish +
 				   bfqq->wr_cur_max_time))) {
-		bfq_log_bfqq(bfqq->bfqd, bfqq,
-		    "resume state: switching off wr");
-
-		bfqq->wr_coeff = 1;
+		if (bfqq->wr_cur_max_time == bfqd->bfq_wr_rt_max_time &&
+		    !bfq_bfqq_in_large_burst(bfqq) &&
+		    time_is_after_eq_jiffies(bfqq->wr_start_at_switch_to_srt +
+					     bfq_wr_duration(bfqd))) {
+			switch_back_to_interactive_wr(bfqq, bfqd);
+		} else {
+			bfqq->wr_coeff = 1;
+			bfq_log_bfqq(bfqq->bfqd, bfqq,
+				     "resume state: switching off wr");
+		}
 	}
 
 	/* make sure weight will be updated, however we got here */
@@ -1173,33 +1218,22 @@ static bool bfq_bfqq_update_budg_for_activation(struct bfq_data *bfqd,
 	return wr_or_deserves_wr;
 }
 
-static unsigned int bfq_wr_duration(struct bfq_data *bfqd)
+/*
+ * Return the farthest future time instant according to jiffies
+ * macros.
+ */
+static unsigned long bfq_greatest_from_now(void)
 {
-	u64 dur;
-
-	if (bfqd->bfq_wr_max_time > 0)
-		return bfqd->bfq_wr_max_time;
-
-	dur = bfqd->RT_prod;
-	do_div(dur, bfqd->peak_rate);
-
-	/*
-	 * Limit duration between 3 and 13 seconds. Tests show that
-	 * higher values than 13 seconds often yield the opposite of
-	 * the desired result, i.e., worsen responsiveness by letting
-	 * non-interactive and non-soft-real-time applications
-	 * preserve weight raising for a too long time interval.
-	 *
-	 * On the other end, lower values than 3 seconds make it
-	 * difficult for most interactive tasks to complete their jobs
-	 * before weight-raising finishes.
-	 */
-	if (dur > msecs_to_jiffies(13000))
-		dur = msecs_to_jiffies(13000);
-	else if (dur < msecs_to_jiffies(3000))
-		dur = msecs_to_jiffies(3000);
+	return jiffies + MAX_JIFFY_OFFSET;
+}
 
-	return dur;
+/*
+ * Return the farthest past time instant according to jiffies
+ * macros.
+ */
+static unsigned long bfq_smallest_from_now(void)
+{
+	return jiffies - MAX_JIFFY_OFFSET;
 }
 
 static void bfq_update_bfqq_wr_on_rq_arrival(struct bfq_data *bfqd,
@@ -1216,7 +1250,19 @@ static void bfq_update_bfqq_wr_on_rq_arrival(struct bfq_data *bfqd,
 			bfqq->wr_coeff = bfqd->bfq_wr_coeff;
 			bfqq->wr_cur_max_time = bfq_wr_duration(bfqd);
 		} else {
-			bfqq->wr_start_at_switch_to_srt = jiffies;
+			/*
+			 * No interactive weight raising in progress
+			 * here: assign minus infinity to
+			 * wr_start_at_switch_to_srt, to make sure
+			 * that, at the end of the soft-real-time
+			 * weight raising periods that is starting
+			 * now, no interactive weight-raising period
+			 * may be wrongly considered as still in
+			 * progress (and thus actually started by
+			 * mistake).
+			 */
+			bfqq->wr_start_at_switch_to_srt =
+				bfq_smallest_from_now();
 			bfqq->wr_coeff = bfqd->bfq_wr_coeff *
 				BFQ_SOFTRT_WEIGHT_FACTOR;
 			bfqq->wr_cur_max_time =
@@ -2016,10 +2062,27 @@ static void bfq_bfqq_save_state(struct bfq_queue *bfqq)
 	bic->saved_IO_bound = bfq_bfqq_IO_bound(bfqq);
 	bic->saved_in_large_burst = bfq_bfqq_in_large_burst(bfqq);
 	bic->was_in_burst_list = !hlist_unhashed(&bfqq->burst_list_node);
-	bic->saved_wr_coeff = bfqq->wr_coeff;
-	bic->saved_wr_start_at_switch_to_srt = bfqq->wr_start_at_switch_to_srt;
-	bic->saved_last_wr_start_finish = bfqq->last_wr_start_finish;
-	bic->saved_wr_cur_max_time = bfqq->wr_cur_max_time;
+	if (unlikely(bfq_bfqq_just_created(bfqq) &&
+		     !bfq_bfqq_in_large_burst(bfqq))) {
+		/*
+		 * bfqq being merged right after being created: bfqq
+		 * would have deserved interactive weight raising, but
+		 * did not make it to be set in a weight-raised state,
+		 * because of this early merge.	Store directly the
+		 * weight-raising state that would have been assigned
+		 * to bfqq, so that to avoid that bfqq unjustly fails
+		 * to enjoy weight raising if split soon.
+		 */
+		bic->saved_wr_coeff = bfqq->bfqd->bfq_wr_coeff;
+		bic->saved_wr_cur_max_time = bfq_wr_duration(bfqq->bfqd);
+		bic->saved_last_wr_start_finish = jiffies;
+	} else {
+		bic->saved_wr_coeff = bfqq->wr_coeff;
+		bic->saved_wr_start_at_switch_to_srt =
+			bfqq->wr_start_at_switch_to_srt;
+		bic->saved_last_wr_start_finish = bfqq->last_wr_start_finish;
+		bic->saved_wr_cur_max_time = bfqq->wr_cur_max_time;
+	}
 }
 
 static void
@@ -2897,24 +2960,6 @@ static unsigned long bfq_bfqq_softrt_next_start(struct bfq_data *bfqd,
 		   jiffies + nsecs_to_jiffies(bfqq->bfqd->bfq_slice_idle) + 4);
 }
 
-/*
- * Return the farthest future time instant according to jiffies
- * macros.
- */
-static unsigned long bfq_greatest_from_now(void)
-{
-	return jiffies + MAX_JIFFY_OFFSET;
-}
-
-/*
- * Return the farthest past time instant according to jiffies
- * macros.
- */
-static unsigned long bfq_smallest_from_now(void)
-{
-	return jiffies - MAX_JIFFY_OFFSET;
-}
-
 /**
  * bfq_bfqq_expire - expire a queue.
  * @bfqd: device owning the queue.
@@ -3489,11 +3534,7 @@ static void bfq_update_wr_data(struct bfq_data *bfqd, struct bfq_queue *bfqq)
 					       bfq_wr_duration(bfqd)))
 				bfq_bfqq_end_wr(bfqq);
 			else {
-				/* switch back to interactive wr */
-				bfqq->wr_coeff = bfqd->bfq_wr_coeff;
-				bfqq->wr_cur_max_time = bfq_wr_duration(bfqd);
-				bfqq->last_wr_start_finish =
-					bfqq->wr_start_at_switch_to_srt;
+				switch_back_to_interactive_wr(bfqq, bfqd);
 				bfqq->entity.prio_changed = 1;
 			}
 		}
@@ -3685,16 +3726,37 @@ void bfq_put_queue(struct bfq_queue *bfqq)
 	if (bfqq->ref)
 		return;
 
-	if (bfq_bfqq_sync(bfqq))
+	if (!hlist_unhashed(&bfqq->burst_list_node)) {
+		hlist_del_init(&bfqq->burst_list_node);
 		/*
-		 * The fact that this queue is being destroyed does not
-		 * invalidate the fact that this queue may have been
-		 * activated during the current burst. As a consequence,
-		 * although the queue does not exist anymore, and hence
-		 * needs to be removed from the burst list if there,
-		 * the burst size has not to be decremented.
+		 * Decrement also burst size after the removal, if the
+		 * process associated with bfqq is exiting, and thus
+		 * does not contribute to the burst any longer. This
+		 * decrement helps filter out false positives of large
+		 * bursts, when some short-lived process (often due to
+		 * the execution of commands by some service) happens
+		 * to start and exit while a complex application is
+		 * starting, and thus spawning several processes that
+		 * do I/O (and that *must not* be treated as a large
+		 * burst, see comments on bfq_handle_burst).
+		 *
+		 * In particular, the decrement is performed only if:
+		 * 1) bfqq is not a merged queue, because, if it is,
+		 * then this free of bfqq is not triggered by the exit
+		 * of the process bfqq is associated with, but exactly
+		 * by the fact that bfqq has just been merged.
+		 * 2) burst_size is greater than 0, to handle
+		 * unbalanced decrements. Unbalanced decrements may
+		 * happen in te following case: bfqq is inserted into
+		 * the current burst list--without incrementing
+		 * bust_size--because of a split, but the current
+		 * burst list is not the burst list bfqq belonged to
+		 * (see comments on the case of a split in
+		 * bfq_set_request).
 		 */
-		hlist_del_init(&bfqq->burst_list_node);
+		if (bfqq->bic && bfqq->bfqd->burst_size > 0)
+			bfqq->bfqd->burst_size--;
+	}
 
 	kmem_cache_free(bfq_pool, bfqq);
 #ifdef CONFIG_BFQ_GROUP_IOSCHED
@@ -4127,7 +4189,6 @@ static void __bfq_insert_request(struct bfq_data *bfqd, struct request *rq)
 		new_bfqq->allocated++;
 		bfqq->allocated--;
 		new_bfqq->ref++;
-		bfq_clear_bfqq_just_created(bfqq);
 		/*
 		 * If the bic associated with the process
 		 * issuing this request still points to bfqq
@@ -4139,6 +4200,8 @@ static void __bfq_insert_request(struct bfq_data *bfqd, struct request *rq)
 		if (bic_to_bfqq(RQ_BIC(rq), 1) == bfqq)
 			bfq_merge_bfqqs(bfqd, RQ_BIC(rq),
 					bfqq, new_bfqq);
+
+		bfq_clear_bfqq_just_created(bfqq);
 		/*
 		 * rq is about to be enqueued into new_bfqq,
 		 * release rq reference on bfqq
@@ -4424,6 +4487,34 @@ static struct bfq_queue *bfq_get_bfqq_handle_split(struct bfq_data *bfqd,
 		else {
 			bfq_clear_bfqq_in_large_burst(bfqq);
 			if (bic->was_in_burst_list)
+				/*
+				 * If bfqq was in the current
+				 * burst list before being
+				 * merged, then we have to add
+				 * it back. And we do not need
+				 * to increase burst_size, as
+				 * we did not decrement
+				 * burst_size when we removed
+				 * bfqq from the burst list as
+				 * a consequence of a merge
+				 * (see comments in
+				 * bfq_put_queue). In this
+				 * respect, it would be rather
+				 * costly to know whether the
+				 * current burst list is still
+				 * the same burst list from
+				 * which bfqq was removed on
+				 * the merge. To avoid this
+				 * cost, if bfqq was in a
+				 * burst list, then we add
+				 * bfqq to the current burst
+				 * list without any further
+				 * check. This can cause
+				 * inappropriate insertions,
+				 * but rarely enough to not
+				 * harm the detection of large
+				 * bursts significantly.
+				 */
 				hlist_add_head(&bfqq->burst_list_node,
 					       &bfqd->burst_list);
 		}
@@ -4775,7 +4866,7 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e)
 	bfq_init_root_group(bfqd->root_group, bfqd);
 	bfq_init_entity(&bfqd->oom_bfqq.entity, bfqd->root_group);
 
-
+	wbt_disable_default(q);
 	return 0;
 
 out_free:
diff --git a/block/bio-integrity.c b/block/bio-integrity.c
index 5df32907ff3b..23b42e8aa03e 100644
--- a/block/bio-integrity.c
+++ b/block/bio-integrity.c
@@ -485,11 +485,8 @@ EXPORT_SYMBOL(bioset_integrity_create);
 
 void bioset_integrity_free(struct bio_set *bs)
 {
-	if (bs->bio_integrity_pool)
-		mempool_destroy(bs->bio_integrity_pool);
-
-	if (bs->bvec_integrity_pool)
-		mempool_destroy(bs->bvec_integrity_pool);
+	mempool_destroy(bs->bio_integrity_pool);
+	mempool_destroy(bs->bvec_integrity_pool);
 }
 EXPORT_SYMBOL(bioset_integrity_free);
 
diff --git a/block/bio.c b/block/bio.c
index cc60213e56d8..b94a802f8ba3 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -400,7 +400,7 @@ static void punt_bios_to_rescuer(struct bio_set *bs)
 
 /**
  * bio_alloc_bioset - allocate a bio for I/O
- * @gfp_mask:   the GFP_ mask given to the slab allocator
+ * @gfp_mask:   the GFP_* mask given to the slab allocator
  * @nr_iovecs:	number of iovecs to pre-allocate
  * @bs:		the bio_set to allocate from.
  *
@@ -1931,11 +1931,8 @@ void bioset_free(struct bio_set *bs)
 	if (bs->rescue_workqueue)
 		destroy_workqueue(bs->rescue_workqueue);
 
-	if (bs->bio_pool)
-		mempool_destroy(bs->bio_pool);
-
-	if (bs->bvec_pool)
-		mempool_destroy(bs->bvec_pool);
+	mempool_destroy(bs->bio_pool);
+	mempool_destroy(bs->bvec_pool);
 
 	bioset_integrity_free(bs);
 	bio_put_slab(bs);
@@ -2036,37 +2033,6 @@ int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css)
 EXPORT_SYMBOL_GPL(bio_associate_blkcg);
 
 /**
- * bio_associate_current - associate a bio with %current
- * @bio: target bio
- *
- * Associate @bio with %current if it hasn't been associated yet.  Block
- * layer will treat @bio as if it were issued by %current no matter which
- * task actually issues it.
- *
- * This function takes an extra reference of @task's io_context and blkcg
- * which will be put when @bio is released.  The caller must own @bio,
- * ensure %current->io_context exists, and is responsible for synchronizing
- * calls to this function.
- */
-int bio_associate_current(struct bio *bio)
-{
-	struct io_context *ioc;
-
-	if (bio->bi_css)
-		return -EBUSY;
-
-	ioc = current->io_context;
-	if (!ioc)
-		return -ENOENT;
-
-	get_io_context_active(ioc);
-	bio->bi_ioc = ioc;
-	bio->bi_css = task_get_css(current, io_cgrp_id);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(bio_associate_current);
-
-/**
  * bio_disassociate_task - undo bio_associate_current()
  * @bio: target bio
  */
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index d3f56baee936..4117524ca45b 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -1419,6 +1419,11 @@ int blkcg_policy_register(struct blkcg_policy *pol)
 	if (i >= BLKCG_MAX_POLS)
 		goto err_unlock;
 
+	/* Make sure cpd/pd_alloc_fn and cpd/pd_free_fn in pairs */
+	if ((!pol->cpd_alloc_fn ^ !pol->cpd_free_fn) ||
+		(!pol->pd_alloc_fn ^ !pol->pd_free_fn))
+		goto err_unlock;
+
 	/* register @pol */
 	pol->plid = i;
 	blkcg_policy[pol->plid] = pol;
@@ -1452,7 +1457,7 @@ int blkcg_policy_register(struct blkcg_policy *pol)
 	return 0;
 
 err_free_cpds:
-	if (pol->cpd_alloc_fn) {
+	if (pol->cpd_free_fn) {
 		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
 			if (blkcg->cpd[pol->plid]) {
 				pol->cpd_free_fn(blkcg->cpd[pol->plid]);
@@ -1492,7 +1497,7 @@ void blkcg_policy_unregister(struct blkcg_policy *pol)
 	/* remove cpds and unregister */
 	mutex_lock(&blkcg_pol_mutex);
 
-	if (pol->cpd_alloc_fn) {
+	if (pol->cpd_free_fn) {
 		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
 			if (blkcg->cpd[pol->plid]) {
 				pol->cpd_free_fn(blkcg->cpd[pol->plid]);
diff --git a/block/blk-core.c b/block/blk-core.c
index 048be4aa6024..7c54c195e79e 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -333,11 +333,13 @@ EXPORT_SYMBOL(blk_stop_queue);
 void blk_sync_queue(struct request_queue *q)
 {
 	del_timer_sync(&q->timeout);
+	cancel_work_sync(&q->timeout_work);
 
 	if (q->mq_ops) {
 		struct blk_mq_hw_ctx *hctx;
 		int i;
 
+		cancel_delayed_work_sync(&q->requeue_work);
 		queue_for_each_hw_ctx(q, hctx, i)
 			cancel_delayed_work_sync(&hctx->run_work);
 	} else {
@@ -347,6 +349,37 @@ void blk_sync_queue(struct request_queue *q)
 EXPORT_SYMBOL(blk_sync_queue);
 
 /**
+ * blk_set_preempt_only - set QUEUE_FLAG_PREEMPT_ONLY
+ * @q: request queue pointer
+ *
+ * Returns the previous value of the PREEMPT_ONLY flag - 0 if the flag was not
+ * set and 1 if the flag was already set.
+ */
+int blk_set_preempt_only(struct request_queue *q)
+{
+	unsigned long flags;
+	int res;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	res = queue_flag_test_and_set(QUEUE_FLAG_PREEMPT_ONLY, q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	return res;
+}
+EXPORT_SYMBOL_GPL(blk_set_preempt_only);
+
+void blk_clear_preempt_only(struct request_queue *q)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	queue_flag_clear(QUEUE_FLAG_PREEMPT_ONLY, q);
+	wake_up_all(&q->mq_freeze_wq);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+EXPORT_SYMBOL_GPL(blk_clear_preempt_only);
+
+/**
  * __blk_run_queue_uncond - run a queue whether or not it has been stopped
  * @q:	The queue to run
  *
@@ -610,6 +643,9 @@ void blk_set_queue_dying(struct request_queue *q)
 		}
 		spin_unlock_irq(q->queue_lock);
 	}
+
+	/* Make blk_queue_enter() reexamine the DYING flag. */
+	wake_up_all(&q->mq_freeze_wq);
 }
 EXPORT_SYMBOL_GPL(blk_set_queue_dying);
 
@@ -718,7 +754,7 @@ static void free_request_size(void *element, void *data)
 int blk_init_rl(struct request_list *rl, struct request_queue *q,
 		gfp_t gfp_mask)
 {
-	if (unlikely(rl->rq_pool))
+	if (unlikely(rl->rq_pool) || q->mq_ops)
 		return 0;
 
 	rl->q = q;
@@ -760,15 +796,38 @@ struct request_queue *blk_alloc_queue(gfp_t gfp_mask)
 }
 EXPORT_SYMBOL(blk_alloc_queue);
 
-int blk_queue_enter(struct request_queue *q, bool nowait)
+/**
+ * blk_queue_enter() - try to increase q->q_usage_counter
+ * @q: request queue pointer
+ * @flags: BLK_MQ_REQ_NOWAIT and/or BLK_MQ_REQ_PREEMPT
+ */
+int blk_queue_enter(struct request_queue *q, blk_mq_req_flags_t flags)
 {
+	const bool preempt = flags & BLK_MQ_REQ_PREEMPT;
+
 	while (true) {
+		bool success = false;
 		int ret;
 
-		if (percpu_ref_tryget_live(&q->q_usage_counter))
+		rcu_read_lock_sched();
+		if (percpu_ref_tryget_live(&q->q_usage_counter)) {
+			/*
+			 * The code that sets the PREEMPT_ONLY flag is
+			 * responsible for ensuring that that flag is globally
+			 * visible before the queue is unfrozen.
+			 */
+			if (preempt || !blk_queue_preempt_only(q)) {
+				success = true;
+			} else {
+				percpu_ref_put(&q->q_usage_counter);
+			}
+		}
+		rcu_read_unlock_sched();
+
+		if (success)
 			return 0;
 
-		if (nowait)
+		if (flags & BLK_MQ_REQ_NOWAIT)
 			return -EBUSY;
 
 		/*
@@ -781,7 +840,8 @@ int blk_queue_enter(struct request_queue *q, bool nowait)
 		smp_rmb();
 
 		ret = wait_event_interruptible(q->mq_freeze_wq,
-				!atomic_read(&q->mq_freeze_depth) ||
+				(atomic_read(&q->mq_freeze_depth) == 0 &&
+				 (preempt || !blk_queue_preempt_only(q))) ||
 				blk_queue_dying(q));
 		if (blk_queue_dying(q))
 			return -ENODEV;
@@ -844,6 +904,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 	setup_timer(&q->backing_dev_info->laptop_mode_wb_timer,
 		    laptop_mode_timer_fn, (unsigned long) q);
 	setup_timer(&q->timeout, blk_rq_timed_out_timer, (unsigned long) q);
+	INIT_WORK(&q->timeout_work, NULL);
 	INIT_LIST_HEAD(&q->queue_head);
 	INIT_LIST_HEAD(&q->timeout_list);
 	INIT_LIST_HEAD(&q->icq_list);
@@ -1154,7 +1215,7 @@ int blk_update_nr_requests(struct request_queue *q, unsigned int nr)
  * @rl: request list to allocate from
  * @op: operation and flags
  * @bio: bio to allocate request for (can be %NULL)
- * @gfp_mask: allocation mask
+ * @flags: BLQ_MQ_REQ_* flags
  *
  * Get a free request from @q.  This function may fail under memory
  * pressure or if @q is dead.
@@ -1164,7 +1225,7 @@ int blk_update_nr_requests(struct request_queue *q, unsigned int nr)
  * Returns request pointer on success, with @q->queue_lock *not held*.
  */
 static struct request *__get_request(struct request_list *rl, unsigned int op,
-		struct bio *bio, gfp_t gfp_mask)
+				     struct bio *bio, blk_mq_req_flags_t flags)
 {
 	struct request_queue *q = rl->q;
 	struct request *rq;
@@ -1173,6 +1234,8 @@ static struct request *__get_request(struct request_list *rl, unsigned int op,
 	struct io_cq *icq = NULL;
 	const bool is_sync = op_is_sync(op);
 	int may_queue;
+	gfp_t gfp_mask = flags & BLK_MQ_REQ_NOWAIT ? GFP_ATOMIC :
+			 __GFP_DIRECT_RECLAIM;
 	req_flags_t rq_flags = RQF_ALLOCED;
 
 	lockdep_assert_held(q->queue_lock);
@@ -1255,6 +1318,8 @@ static struct request *__get_request(struct request_list *rl, unsigned int op,
 	blk_rq_set_rl(rq, rl);
 	rq->cmd_flags = op;
 	rq->rq_flags = rq_flags;
+	if (flags & BLK_MQ_REQ_PREEMPT)
+		rq->rq_flags |= RQF_PREEMPT;
 
 	/* init elvpriv */
 	if (rq_flags & RQF_ELVPRIV) {
@@ -1333,7 +1398,7 @@ rq_starved:
  * @q: request_queue to allocate request from
  * @op: operation and flags
  * @bio: bio to allocate request for (can be %NULL)
- * @gfp_mask: allocation mask
+ * @flags: BLK_MQ_REQ_* flags.
  *
  * Get a free request from @q.  If %__GFP_DIRECT_RECLAIM is set in @gfp_mask,
  * this function keeps retrying under memory pressure and fails iff @q is dead.
@@ -1343,7 +1408,7 @@ rq_starved:
  * Returns request pointer on success, with @q->queue_lock *not held*.
  */
 static struct request *get_request(struct request_queue *q, unsigned int op,
-		struct bio *bio, gfp_t gfp_mask)
+				   struct bio *bio, blk_mq_req_flags_t flags)
 {
 	const bool is_sync = op_is_sync(op);
 	DEFINE_WAIT(wait);
@@ -1355,7 +1420,7 @@ static struct request *get_request(struct request_queue *q, unsigned int op,
 
 	rl = blk_get_rl(q, bio);	/* transferred to @rq on success */
 retry:
-	rq = __get_request(rl, op, bio, gfp_mask);
+	rq = __get_request(rl, op, bio, flags);
 	if (!IS_ERR(rq))
 		return rq;
 
@@ -1364,7 +1429,7 @@ retry:
 		return ERR_PTR(-EAGAIN);
 	}
 
-	if (!gfpflags_allow_blocking(gfp_mask) || unlikely(blk_queue_dying(q))) {
+	if ((flags & BLK_MQ_REQ_NOWAIT) || unlikely(blk_queue_dying(q))) {
 		blk_put_rl(rl);
 		return rq;
 	}
@@ -1391,20 +1456,28 @@ retry:
 	goto retry;
 }
 
+/* flags: BLK_MQ_REQ_PREEMPT and/or BLK_MQ_REQ_NOWAIT. */
 static struct request *blk_old_get_request(struct request_queue *q,
-					   unsigned int op, gfp_t gfp_mask)
+				unsigned int op, blk_mq_req_flags_t flags)
 {
 	struct request *rq;
+	gfp_t gfp_mask = flags & BLK_MQ_REQ_NOWAIT ? GFP_ATOMIC :
+			 __GFP_DIRECT_RECLAIM;
+	int ret = 0;
 
 	WARN_ON_ONCE(q->mq_ops);
 
 	/* create ioc upfront */
 	create_io_context(gfp_mask, q->node);
 
+	ret = blk_queue_enter(q, flags);
+	if (ret)
+		return ERR_PTR(ret);
 	spin_lock_irq(q->queue_lock);
-	rq = get_request(q, op, NULL, gfp_mask);
+	rq = get_request(q, op, NULL, flags);
 	if (IS_ERR(rq)) {
 		spin_unlock_irq(q->queue_lock);
+		blk_queue_exit(q);
 		return rq;
 	}
 
@@ -1415,25 +1488,40 @@ static struct request *blk_old_get_request(struct request_queue *q,
 	return rq;
 }
 
-struct request *blk_get_request(struct request_queue *q, unsigned int op,
-				gfp_t gfp_mask)
+/**
+ * blk_get_request_flags - allocate a request
+ * @q: request queue to allocate a request for
+ * @op: operation (REQ_OP_*) and REQ_* flags, e.g. REQ_SYNC.
+ * @flags: BLK_MQ_REQ_* flags, e.g. BLK_MQ_REQ_NOWAIT.
+ */
+struct request *blk_get_request_flags(struct request_queue *q, unsigned int op,
+				      blk_mq_req_flags_t flags)
 {
 	struct request *req;
 
+	WARN_ON_ONCE(op & REQ_NOWAIT);
+	WARN_ON_ONCE(flags & ~(BLK_MQ_REQ_NOWAIT | BLK_MQ_REQ_PREEMPT));
+
 	if (q->mq_ops) {
-		req = blk_mq_alloc_request(q, op,
-			(gfp_mask & __GFP_DIRECT_RECLAIM) ?
-				0 : BLK_MQ_REQ_NOWAIT);
+		req = blk_mq_alloc_request(q, op, flags);
 		if (!IS_ERR(req) && q->mq_ops->initialize_rq_fn)
 			q->mq_ops->initialize_rq_fn(req);
 	} else {
-		req = blk_old_get_request(q, op, gfp_mask);
+		req = blk_old_get_request(q, op, flags);
 		if (!IS_ERR(req) && q->initialize_rq_fn)
 			q->initialize_rq_fn(req);
 	}
 
 	return req;
 }
+EXPORT_SYMBOL(blk_get_request_flags);
+
+struct request *blk_get_request(struct request_queue *q, unsigned int op,
+				gfp_t gfp_mask)
+{
+	return blk_get_request_flags(q, op, gfp_mask & __GFP_DIRECT_RECLAIM ?
+				     0 : BLK_MQ_REQ_NOWAIT);
+}
 EXPORT_SYMBOL(blk_get_request);
 
 /**
@@ -1576,6 +1664,7 @@ void __blk_put_request(struct request_queue *q, struct request *req)
 		blk_free_request(rl, req);
 		freed_request(rl, sync, rq_flags);
 		blk_put_rl(rl);
+		blk_queue_exit(q);
 	}
 }
 EXPORT_SYMBOL_GPL(__blk_put_request);
@@ -1857,8 +1946,10 @@ get_rq:
 	 * Grab a free request. This is might sleep but can not fail.
 	 * Returns with the queue unlocked.
 	 */
-	req = get_request(q, bio->bi_opf, bio, GFP_NOIO);
+	blk_queue_enter_live(q);
+	req = get_request(q, bio->bi_opf, bio, 0);
 	if (IS_ERR(req)) {
+		blk_queue_exit(q);
 		__wbt_done(q->rq_wb, wb_acct);
 		if (PTR_ERR(req) == -ENOMEM)
 			bio->bi_status = BLK_STS_RESOURCE;
@@ -2200,8 +2291,10 @@ blk_qc_t generic_make_request(struct bio *bio)
 	current->bio_list = bio_list_on_stack;
 	do {
 		struct request_queue *q = bio->bi_disk->queue;
+		blk_mq_req_flags_t flags = bio->bi_opf & REQ_NOWAIT ?
+			BLK_MQ_REQ_NOWAIT : 0;
 
-		if (likely(blk_queue_enter(q, bio->bi_opf & REQ_NOWAIT) == 0)) {
+		if (likely(blk_queue_enter(q, flags) == 0)) {
 			struct bio_list lower, same;
 
 			/* Create a fresh bio_list for all subordinate requests */
@@ -2242,6 +2335,40 @@ out:
 EXPORT_SYMBOL(generic_make_request);
 
 /**
+ * direct_make_request - hand a buffer directly to its device driver for I/O
+ * @bio:  The bio describing the location in memory and on the device.
+ *
+ * This function behaves like generic_make_request(), but does not protect
+ * against recursion.  Must only be used if the called driver is known
+ * to not call generic_make_request (or direct_make_request) again from
+ * its make_request function.  (Calling direct_make_request again from
+ * a workqueue is perfectly fine as that doesn't recurse).
+ */
+blk_qc_t direct_make_request(struct bio *bio)
+{
+	struct request_queue *q = bio->bi_disk->queue;
+	bool nowait = bio->bi_opf & REQ_NOWAIT;
+	blk_qc_t ret;
+
+	if (!generic_make_request_checks(bio))
+		return BLK_QC_T_NONE;
+
+	if (unlikely(blk_queue_enter(q, nowait ? BLK_MQ_REQ_NOWAIT : 0))) {
+		if (nowait && !blk_queue_dying(q))
+			bio->bi_status = BLK_STS_AGAIN;
+		else
+			bio->bi_status = BLK_STS_IOERR;
+		bio_endio(bio);
+		return BLK_QC_T_NONE;
+	}
+
+	ret = q->make_request_fn(q, bio);
+	blk_queue_exit(q);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(direct_make_request);
+
+/**
  * submit_bio - submit a bio to the block device layer for I/O
  * @bio: The &struct bio which describes the I/O
  *
@@ -2285,6 +2412,17 @@ blk_qc_t submit_bio(struct bio *bio)
 }
 EXPORT_SYMBOL(submit_bio);
 
+bool blk_poll(struct request_queue *q, blk_qc_t cookie)
+{
+	if (!q->poll_fn || !blk_qc_t_valid(cookie))
+		return false;
+
+	if (current->plug)
+		blk_flush_plug_list(current->plug, false);
+	return q->poll_fn(q, cookie);
+}
+EXPORT_SYMBOL_GPL(blk_poll);
+
 /**
  * blk_cloned_rq_check_limits - Helper function to check a cloned request
  *                              for new the queue limits
@@ -2350,7 +2488,7 @@ blk_status_t blk_insert_cloned_request(struct request_queue *q, struct request *
 		 * bypass a potential scheduler on the bottom device for
 		 * insert.
 		 */
-		blk_mq_request_bypass_insert(rq);
+		blk_mq_request_bypass_insert(rq, true);
 		return BLK_STS_OK;
 	}
 
@@ -2464,20 +2602,22 @@ void blk_account_io_done(struct request *req)
  * Don't process normal requests when queue is suspended
  * or in the process of suspending/resuming
  */
-static struct request *blk_pm_peek_request(struct request_queue *q,
-					   struct request *rq)
+static bool blk_pm_allow_request(struct request *rq)
 {
-	if (q->dev && (q->rpm_status == RPM_SUSPENDED ||
-	    (q->rpm_status != RPM_ACTIVE && !(rq->rq_flags & RQF_PM))))
-		return NULL;
-	else
-		return rq;
+	switch (rq->q->rpm_status) {
+	case RPM_RESUMING:
+	case RPM_SUSPENDING:
+		return rq->rq_flags & RQF_PM;
+	case RPM_SUSPENDED:
+		return false;
+	}
+
+	return true;
 }
 #else
-static inline struct request *blk_pm_peek_request(struct request_queue *q,
-						  struct request *rq)
+static bool blk_pm_allow_request(struct request *rq)
 {
-	return rq;
+	return true;
 }
 #endif
 
@@ -2517,6 +2657,48 @@ void blk_account_io_start(struct request *rq, bool new_io)
 	part_stat_unlock();
 }
 
+static struct request *elv_next_request(struct request_queue *q)
+{
+	struct request *rq;
+	struct blk_flush_queue *fq = blk_get_flush_queue(q, NULL);
+
+	WARN_ON_ONCE(q->mq_ops);
+
+	while (1) {
+		list_for_each_entry(rq, &q->queue_head, queuelist) {
+			if (blk_pm_allow_request(rq))
+				return rq;
+
+			if (rq->rq_flags & RQF_SOFTBARRIER)
+				break;
+		}
+
+		/*
+		 * Flush request is running and flush request isn't queueable
+		 * in the drive, we can hold the queue till flush request is
+		 * finished. Even we don't do this, driver can't dispatch next
+		 * requests and will requeue them. And this can improve
+		 * throughput too. For example, we have request flush1, write1,
+		 * flush 2. flush1 is dispatched, then queue is hold, write1
+		 * isn't inserted to queue. After flush1 is finished, flush2
+		 * will be dispatched. Since disk cache is already clean,
+		 * flush2 will be finished very soon, so looks like flush2 is
+		 * folded to flush1.
+		 * Since the queue is hold, a flag is set to indicate the queue
+		 * should be restarted later. Please see flush_end_io() for
+		 * details.
+		 */
+		if (fq->flush_pending_idx != fq->flush_running_idx &&
+				!queue_flush_queueable(q)) {
+			fq->flush_queue_delayed = 1;
+			return NULL;
+		}
+		if (unlikely(blk_queue_bypass(q)) ||
+		    !q->elevator->type->ops.sq.elevator_dispatch_fn(q, 0))
+			return NULL;
+	}
+}
+
 /**
  * blk_peek_request - peek at the top of a request queue
  * @q: request queue to peek at
@@ -2538,12 +2720,7 @@ struct request *blk_peek_request(struct request_queue *q)
 	lockdep_assert_held(q->queue_lock);
 	WARN_ON_ONCE(q->mq_ops);
 
-	while ((rq = __elv_next_request(q)) != NULL) {
-
-		rq = blk_pm_peek_request(q, rq);
-		if (!rq)
-			break;
-
+	while ((rq = elv_next_request(q)) != NULL) {
 		if (!(rq->rq_flags & RQF_STARTED)) {
 			/*
 			 * This is the first time the device driver
@@ -2695,6 +2872,27 @@ struct request *blk_fetch_request(struct request_queue *q)
 }
 EXPORT_SYMBOL(blk_fetch_request);
 
+/*
+ * Steal bios from a request and add them to a bio list.
+ * The request must not have been partially completed before.
+ */
+void blk_steal_bios(struct bio_list *list, struct request *rq)
+{
+	if (rq->bio) {
+		if (list->tail)
+			list->tail->bi_next = rq->bio;
+		else
+			list->head = rq->bio;
+		list->tail = rq->biotail;
+
+		rq->bio = NULL;
+		rq->biotail = NULL;
+	}
+
+	rq->__data_len = 0;
+}
+EXPORT_SYMBOL_GPL(blk_steal_bios);
+
 /**
  * blk_update_request - Special helper function for request stacking drivers
  * @req:      the request being processed
diff --git a/block/blk-flush.c b/block/blk-flush.c
index 4938bec8cfef..f17170675917 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -231,8 +231,13 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error)
 		/* release the tag's ownership to the req cloned from */
 		spin_lock_irqsave(&fq->mq_flush_lock, flags);
 		hctx = blk_mq_map_queue(q, flush_rq->mq_ctx->cpu);
-		blk_mq_tag_set_rq(hctx, flush_rq->tag, fq->orig_rq);
-		flush_rq->tag = -1;
+		if (!q->elevator) {
+			blk_mq_tag_set_rq(hctx, flush_rq->tag, fq->orig_rq);
+			flush_rq->tag = -1;
+		} else {
+			blk_mq_put_driver_tag_hctx(hctx, flush_rq);
+			flush_rq->internal_tag = -1;
+		}
 	}
 
 	running = &fq->flush_queue[fq->flush_running_idx];
@@ -318,19 +323,26 @@ static bool blk_kick_flush(struct request_queue *q, struct blk_flush_queue *fq)
 	blk_rq_init(q, flush_rq);
 
 	/*
-	 * Borrow tag from the first request since they can't
-	 * be in flight at the same time. And acquire the tag's
-	 * ownership for flush req.
+	 * In case of none scheduler, borrow tag from the first request
+	 * since they can't be in flight at the same time. And acquire
+	 * the tag's ownership for flush req.
+	 *
+	 * In case of IO scheduler, flush rq need to borrow scheduler tag
+	 * just for cheating put/get driver tag.
 	 */
 	if (q->mq_ops) {
 		struct blk_mq_hw_ctx *hctx;
 
 		flush_rq->mq_ctx = first_rq->mq_ctx;
-		flush_rq->tag = first_rq->tag;
-		fq->orig_rq = first_rq;
 
-		hctx = blk_mq_map_queue(q, first_rq->mq_ctx->cpu);
-		blk_mq_tag_set_rq(hctx, first_rq->tag, flush_rq);
+		if (!q->elevator) {
+			fq->orig_rq = first_rq;
+			flush_rq->tag = first_rq->tag;
+			hctx = blk_mq_map_queue(q, first_rq->mq_ctx->cpu);
+			blk_mq_tag_set_rq(hctx, first_rq->tag, flush_rq);
+		} else {
+			flush_rq->internal_tag = first_rq->internal_tag;
+		}
 	}
 
 	flush_rq->cmd_flags = REQ_OP_FLUSH | REQ_PREFLUSH;
@@ -394,6 +406,11 @@ static void mq_flush_data_end_io(struct request *rq, blk_status_t error)
 
 	hctx = blk_mq_map_queue(q, ctx->cpu);
 
+	if (q->elevator) {
+		WARN_ON(rq->tag < 0);
+		blk_mq_put_driver_tag_hctx(hctx, rq);
+	}
+
 	/*
 	 * After populating an empty queue, kick it to avoid stall.  Read
 	 * the comment in flush_end_io().
@@ -463,7 +480,7 @@ void blk_insert_flush(struct request *rq)
 	if ((policy & REQ_FSEQ_DATA) &&
 	    !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) {
 		if (q->mq_ops)
-			blk_mq_sched_insert_request(rq, false, true, false, false);
+			blk_mq_request_bypass_insert(rq, false);
 		else
 			list_add_tail(&rq->queuelist, &q->queue_head);
 		return;
diff --git a/block/blk-lib.c b/block/blk-lib.c
index 63fb971d6574..2bc544ce3d2e 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -275,6 +275,40 @@ static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
 	return min(pages, (sector_t)BIO_MAX_PAGES);
 }
 
+static int __blkdev_issue_zero_pages(struct block_device *bdev,
+		sector_t sector, sector_t nr_sects, gfp_t gfp_mask,
+		struct bio **biop)
+{
+	struct request_queue *q = bdev_get_queue(bdev);
+	struct bio *bio = *biop;
+	int bi_size = 0;
+	unsigned int sz;
+
+	if (!q)
+		return -ENXIO;
+
+	while (nr_sects != 0) {
+		bio = next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
+			       gfp_mask);
+		bio->bi_iter.bi_sector = sector;
+		bio_set_dev(bio, bdev);
+		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+
+		while (nr_sects != 0) {
+			sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
+			bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
+			nr_sects -= bi_size >> 9;
+			sector += bi_size >> 9;
+			if (bi_size < sz)
+				break;
+		}
+		cond_resched();
+	}
+
+	*biop = bio;
+	return 0;
+}
+
 /**
  * __blkdev_issue_zeroout - generate number of zero filed write bios
  * @bdev:	blockdev to issue
@@ -288,12 +322,6 @@ static unsigned int __blkdev_sectors_to_bio_pages(sector_t nr_sects)
  *  Zero-fill a block range, either using hardware offload or by explicitly
  *  writing zeroes to the device.
  *
- *  Note that this function may fail with -EOPNOTSUPP if the driver signals
- *  zeroing offload support, but the device fails to process the command (for
- *  some devices there is no non-destructive way to verify whether this
- *  operation is actually supported).  In this case the caller should call
- *  retry the call to blkdev_issue_zeroout() and the fallback path will be used.
- *
  *  If a device is using logical block provisioning, the underlying space will
  *  not be released if %flags contains BLKDEV_ZERO_NOUNMAP.
  *
@@ -305,9 +333,6 @@ int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 		unsigned flags)
 {
 	int ret;
-	int bi_size = 0;
-	struct bio *bio = *biop;
-	unsigned int sz;
 	sector_t bs_mask;
 
 	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
@@ -317,30 +342,10 @@ int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 	ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask,
 			biop, flags);
 	if (ret != -EOPNOTSUPP || (flags & BLKDEV_ZERO_NOFALLBACK))
-		goto out;
-
-	ret = 0;
-	while (nr_sects != 0) {
-		bio = next_bio(bio, __blkdev_sectors_to_bio_pages(nr_sects),
-			       gfp_mask);
-		bio->bi_iter.bi_sector = sector;
-		bio_set_dev(bio, bdev);
-		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
-
-		while (nr_sects != 0) {
-			sz = min((sector_t) PAGE_SIZE, nr_sects << 9);
-			bi_size = bio_add_page(bio, ZERO_PAGE(0), sz, 0);
-			nr_sects -= bi_size >> 9;
-			sector += bi_size >> 9;
-			if (bi_size < sz)
-				break;
-		}
-		cond_resched();
-	}
+		return ret;
 
-	*biop = bio;
-out:
-	return ret;
+	return __blkdev_issue_zero_pages(bdev, sector, nr_sects, gfp_mask,
+					 biop);
 }
 EXPORT_SYMBOL(__blkdev_issue_zeroout);
 
@@ -360,18 +365,49 @@ EXPORT_SYMBOL(__blkdev_issue_zeroout);
 int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 		sector_t nr_sects, gfp_t gfp_mask, unsigned flags)
 {
-	int ret;
-	struct bio *bio = NULL;
+	int ret = 0;
+	sector_t bs_mask;
+	struct bio *bio;
 	struct blk_plug plug;
+	bool try_write_zeroes = !!bdev_write_zeroes_sectors(bdev);
 
+	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
+	if ((sector | nr_sects) & bs_mask)
+		return -EINVAL;
+
+retry:
+	bio = NULL;
 	blk_start_plug(&plug);
-	ret = __blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask,
-			&bio, flags);
+	if (try_write_zeroes) {
+		ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects,
+						  gfp_mask, &bio, flags);
+	} else if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
+		ret = __blkdev_issue_zero_pages(bdev, sector, nr_sects,
+						gfp_mask, &bio);
+	} else {
+		/* No zeroing offload support */
+		ret = -EOPNOTSUPP;
+	}
 	if (ret == 0 && bio) {
 		ret = submit_bio_wait(bio);
 		bio_put(bio);
 	}
 	blk_finish_plug(&plug);
+	if (ret && try_write_zeroes) {
+		if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
+			try_write_zeroes = false;
+			goto retry;
+		}
+		if (!bdev_write_zeroes_sectors(bdev)) {
+			/*
+			 * Zeroing offload support was indicated, but the
+			 * device reported ILLEGAL REQUEST (for some devices
+			 * there is no non-destructive way to verify whether
+			 * WRITE ZEROES is actually supported).
+			 */
+			ret = -EOPNOTSUPP;
+		}
+	}
 
 	return ret;
 }
diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c
index de294d775acf..b56a4f35720d 100644
--- a/block/blk-mq-debugfs.c
+++ b/block/blk-mq-debugfs.c
@@ -54,7 +54,6 @@ static const char *const blk_queue_flag_name[] = {
 	QUEUE_FLAG_NAME(NOMERGES),
 	QUEUE_FLAG_NAME(SAME_COMP),
 	QUEUE_FLAG_NAME(FAIL_IO),
-	QUEUE_FLAG_NAME(STACKABLE),
 	QUEUE_FLAG_NAME(NONROT),
 	QUEUE_FLAG_NAME(IO_STAT),
 	QUEUE_FLAG_NAME(DISCARD),
@@ -75,6 +74,7 @@ static const char *const blk_queue_flag_name[] = {
 	QUEUE_FLAG_NAME(REGISTERED),
 	QUEUE_FLAG_NAME(SCSI_PASSTHROUGH),
 	QUEUE_FLAG_NAME(QUIESCED),
+	QUEUE_FLAG_NAME(PREEMPT_ONLY),
 };
 #undef QUEUE_FLAG_NAME
 
@@ -180,7 +180,6 @@ static const char *const hctx_state_name[] = {
 	HCTX_STATE_NAME(STOPPED),
 	HCTX_STATE_NAME(TAG_ACTIVE),
 	HCTX_STATE_NAME(SCHED_RESTART),
-	HCTX_STATE_NAME(TAG_WAITING),
 	HCTX_STATE_NAME(START_ON_RUN),
 };
 #undef HCTX_STATE_NAME
diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index 4ab69435708c..c117bd8fd1f6 100644
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -81,20 +81,103 @@ static bool blk_mq_sched_restart_hctx(struct blk_mq_hw_ctx *hctx)
 	} else
 		clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);
 
-	if (blk_mq_hctx_has_pending(hctx)) {
-		blk_mq_run_hw_queue(hctx, true);
-		return true;
-	}
+	return blk_mq_run_hw_queue(hctx, true);
+}
 
-	return false;
+/*
+ * Only SCSI implements .get_budget and .put_budget, and SCSI restarts
+ * its queue by itself in its completion handler, so we don't need to
+ * restart queue if .get_budget() returns BLK_STS_NO_RESOURCE.
+ */
+static void blk_mq_do_dispatch_sched(struct blk_mq_hw_ctx *hctx)
+{
+	struct request_queue *q = hctx->queue;
+	struct elevator_queue *e = q->elevator;
+	LIST_HEAD(rq_list);
+
+	do {
+		struct request *rq;
+
+		if (e->type->ops.mq.has_work &&
+				!e->type->ops.mq.has_work(hctx))
+			break;
+
+		if (!blk_mq_get_dispatch_budget(hctx))
+			break;
+
+		rq = e->type->ops.mq.dispatch_request(hctx);
+		if (!rq) {
+			blk_mq_put_dispatch_budget(hctx);
+			break;
+		}
+
+		/*
+		 * Now this rq owns the budget which has to be released
+		 * if this rq won't be queued to driver via .queue_rq()
+		 * in blk_mq_dispatch_rq_list().
+		 */
+		list_add(&rq->queuelist, &rq_list);
+	} while (blk_mq_dispatch_rq_list(q, &rq_list, true));
 }
 
+static struct blk_mq_ctx *blk_mq_next_ctx(struct blk_mq_hw_ctx *hctx,
+					  struct blk_mq_ctx *ctx)
+{
+	unsigned idx = ctx->index_hw;
+
+	if (++idx == hctx->nr_ctx)
+		idx = 0;
+
+	return hctx->ctxs[idx];
+}
+
+/*
+ * Only SCSI implements .get_budget and .put_budget, and SCSI restarts
+ * its queue by itself in its completion handler, so we don't need to
+ * restart queue if .get_budget() returns BLK_STS_NO_RESOURCE.
+ */
+static void blk_mq_do_dispatch_ctx(struct blk_mq_hw_ctx *hctx)
+{
+	struct request_queue *q = hctx->queue;
+	LIST_HEAD(rq_list);
+	struct blk_mq_ctx *ctx = READ_ONCE(hctx->dispatch_from);
+
+	do {
+		struct request *rq;
+
+		if (!sbitmap_any_bit_set(&hctx->ctx_map))
+			break;
+
+		if (!blk_mq_get_dispatch_budget(hctx))
+			break;
+
+		rq = blk_mq_dequeue_from_ctx(hctx, ctx);
+		if (!rq) {
+			blk_mq_put_dispatch_budget(hctx);
+			break;
+		}
+
+		/*
+		 * Now this rq owns the budget which has to be released
+		 * if this rq won't be queued to driver via .queue_rq()
+		 * in blk_mq_dispatch_rq_list().
+		 */
+		list_add(&rq->queuelist, &rq_list);
+
+		/* round robin for fair dispatch */
+		ctx = blk_mq_next_ctx(hctx, rq->mq_ctx);
+
+	} while (blk_mq_dispatch_rq_list(q, &rq_list, true));
+
+	WRITE_ONCE(hctx->dispatch_from, ctx);
+}
+
+/* return true if hw queue need to be run again */
 void blk_mq_sched_dispatch_requests(struct blk_mq_hw_ctx *hctx)
 {
 	struct request_queue *q = hctx->queue;
 	struct elevator_queue *e = q->elevator;
 	const bool has_sched_dispatch = e && e->type->ops.mq.dispatch_request;
-	bool did_work = false;
 	LIST_HEAD(rq_list);
 
 	/* RCU or SRCU read lock is needed before checking quiesced flag */
@@ -122,29 +205,34 @@ void blk_mq_sched_dispatch_requests(struct blk_mq_hw_ctx *hctx)
 	 * scheduler, we can no longer merge or sort them. So it's best to
 	 * leave them there for as long as we can. Mark the hw queue as
 	 * needing a restart in that case.
+	 *
+	 * We want to dispatch from the scheduler if there was nothing
+	 * on the dispatch list or we were able to dispatch from the
+	 * dispatch list.
 	 */
 	if (!list_empty(&rq_list)) {
 		blk_mq_sched_mark_restart_hctx(hctx);
-		did_work = blk_mq_dispatch_rq_list(q, &rq_list);
-	} else if (!has_sched_dispatch) {
+		if (blk_mq_dispatch_rq_list(q, &rq_list, false)) {
+			if (has_sched_dispatch)
+				blk_mq_do_dispatch_sched(hctx);
+			else
+				blk_mq_do_dispatch_ctx(hctx);
+		}
+	} else if (has_sched_dispatch) {
+		blk_mq_do_dispatch_sched(hctx);
+	} else if (q->mq_ops->get_budget) {
+		/*
+		 * If we need to get budget before queuing request, we
+		 * dequeue request one by one from sw queue for avoiding
+		 * to mess up I/O merge when dispatch runs out of resource.
+		 *
+		 * TODO: get more budgets, and dequeue more requests in
+		 * one time.
+		 */
+		blk_mq_do_dispatch_ctx(hctx);
+	} else {
 		blk_mq_flush_busy_ctxs(hctx, &rq_list);
-		blk_mq_dispatch_rq_list(q, &rq_list);
-	}
-
-	/*
-	 * We want to dispatch from the scheduler if we had no work left
-	 * on the dispatch list, OR if we did have work but weren't able
-	 * to make progress.
-	 */
-	if (!did_work && has_sched_dispatch) {
-		do {
-			struct request *rq;
-
-			rq = e->type->ops.mq.dispatch_request(hctx);
-			if (!rq)
-				break;
-			list_add(&rq->queuelist, &rq_list);
-		} while (blk_mq_dispatch_rq_list(q, &rq_list));
+		blk_mq_dispatch_rq_list(q, &rq_list, false);
 	}
 }
 
@@ -260,21 +348,21 @@ void blk_mq_sched_request_inserted(struct request *rq)
 EXPORT_SYMBOL_GPL(blk_mq_sched_request_inserted);
 
 static bool blk_mq_sched_bypass_insert(struct blk_mq_hw_ctx *hctx,
+				       bool has_sched,
 				       struct request *rq)
 {
-	if (rq->tag == -1) {
-		rq->rq_flags |= RQF_SORTED;
-		return false;
+	/* dispatch flush rq directly */
+	if (rq->rq_flags & RQF_FLUSH_SEQ) {
+		spin_lock(&hctx->lock);
+		list_add(&rq->queuelist, &hctx->dispatch);
+		spin_unlock(&hctx->lock);
+		return true;
 	}
 
-	/*
-	 * If we already have a real request tag, send directly to
-	 * the dispatch list.
-	 */
-	spin_lock(&hctx->lock);
-	list_add(&rq->queuelist, &hctx->dispatch);
-	spin_unlock(&hctx->lock);
-	return true;
+	if (has_sched)
+		rq->rq_flags |= RQF_SORTED;
+
+	return false;
 }
 
 /**
@@ -339,21 +427,6 @@ done:
 	}
 }
 
-/*
- * Add flush/fua to the queue. If we fail getting a driver tag, then
- * punt to the requeue list. Requeue will re-invoke us from a context
- * that's safe to block from.
- */
-static void blk_mq_sched_insert_flush(struct blk_mq_hw_ctx *hctx,
-				      struct request *rq, bool can_block)
-{
-	if (blk_mq_get_driver_tag(rq, &hctx, can_block)) {
-		blk_insert_flush(rq);
-		blk_mq_run_hw_queue(hctx, true);
-	} else
-		blk_mq_add_to_requeue_list(rq, false, true);
-}
-
 void blk_mq_sched_insert_request(struct request *rq, bool at_head,
 				 bool run_queue, bool async, bool can_block)
 {
@@ -362,12 +435,15 @@ void blk_mq_sched_insert_request(struct request *rq, bool at_head,
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 	struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, ctx->cpu);
 
-	if (rq->tag == -1 && op_is_flush(rq->cmd_flags)) {
-		blk_mq_sched_insert_flush(hctx, rq, can_block);
-		return;
+	/* flush rq in flush machinery need to be dispatched directly */
+	if (!(rq->rq_flags & RQF_FLUSH_SEQ) && op_is_flush(rq->cmd_flags)) {
+		blk_insert_flush(rq);
+		goto run;
 	}
 
-	if (e && blk_mq_sched_bypass_insert(hctx, rq))
+	WARN_ON(e && (rq->tag != -1));
+
+	if (blk_mq_sched_bypass_insert(hctx, !!e, rq))
 		goto run;
 
 	if (e && e->type->ops.mq.insert_requests) {
@@ -393,23 +469,6 @@ void blk_mq_sched_insert_requests(struct request_queue *q,
 	struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, ctx->cpu);
 	struct elevator_queue *e = hctx->queue->elevator;
 
-	if (e) {
-		struct request *rq, *next;
-
-		/*
-		 * We bypass requests that already have a driver tag assigned,
-		 * which should only be flushes. Flushes are only ever inserted
-		 * as single requests, so we shouldn't ever hit the
-		 * WARN_ON_ONCE() below (but let's handle it just in case).
-		 */
-		list_for_each_entry_safe(rq, next, list, queuelist) {
-			if (WARN_ON_ONCE(rq->tag != -1)) {
-				list_del_init(&rq->queuelist);
-				blk_mq_sched_bypass_insert(hctx, rq);
-			}
-		}
-	}
-
 	if (e && e->type->ops.mq.insert_requests)
 		e->type->ops.mq.insert_requests(hctx, list, false);
 	else
diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index 6714507aa6c7..c81b40ecd3f1 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -298,12 +298,12 @@ void blk_mq_tagset_busy_iter(struct blk_mq_tag_set *tagset,
 }
 EXPORT_SYMBOL(blk_mq_tagset_busy_iter);
 
-int blk_mq_reinit_tagset(struct blk_mq_tag_set *set,
-			 int (reinit_request)(void *, struct request *))
+int blk_mq_tagset_iter(struct blk_mq_tag_set *set, void *data,
+			 int (fn)(void *, struct request *))
 {
 	int i, j, ret = 0;
 
-	if (WARN_ON_ONCE(!reinit_request))
+	if (WARN_ON_ONCE(!fn))
 		goto out;
 
 	for (i = 0; i < set->nr_hw_queues; i++) {
@@ -316,8 +316,7 @@ int blk_mq_reinit_tagset(struct blk_mq_tag_set *set,
 			if (!tags->static_rqs[j])
 				continue;
 
-			ret = reinit_request(set->driver_data,
-					     tags->static_rqs[j]);
+			ret = fn(data, tags->static_rqs[j]);
 			if (ret)
 				goto out;
 		}
@@ -326,7 +325,7 @@ int blk_mq_reinit_tagset(struct blk_mq_tag_set *set,
 out:
 	return ret;
 }
-EXPORT_SYMBOL_GPL(blk_mq_reinit_tagset);
+EXPORT_SYMBOL_GPL(blk_mq_tagset_iter);
 
 void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn,
 		void *priv)
diff --git a/block/blk-mq-tag.h b/block/blk-mq-tag.h
index c190165d92ea..61deab0b5a5a 100644
--- a/block/blk-mq-tag.h
+++ b/block/blk-mq-tag.h
@@ -45,13 +45,8 @@ static inline struct sbq_wait_state *bt_wait_ptr(struct sbitmap_queue *bt,
 }
 
 enum {
-	BLK_MQ_TAG_CACHE_MIN	= 1,
-	BLK_MQ_TAG_CACHE_MAX	= 64,
-};
-
-enum {
 	BLK_MQ_TAG_FAIL		= -1U,
-	BLK_MQ_TAG_MIN		= BLK_MQ_TAG_CACHE_MIN,
+	BLK_MQ_TAG_MIN		= 1,
 	BLK_MQ_TAG_MAX		= BLK_MQ_TAG_FAIL - 1,
 };
 
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 98a18609755e..b600463791ec 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -37,6 +37,7 @@
 #include "blk-wbt.h"
 #include "blk-mq-sched.h"
 
+static bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie);
 static void blk_mq_poll_stats_start(struct request_queue *q);
 static void blk_mq_poll_stats_fn(struct blk_stat_callback *cb);
 
@@ -60,10 +61,10 @@ static int blk_mq_poll_stats_bkt(const struct request *rq)
 /*
  * Check if any of the ctx's have pending work in this hardware queue
  */
-bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
+static bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
 {
-	return sbitmap_any_bit_set(&hctx->ctx_map) ||
-			!list_empty_careful(&hctx->dispatch) ||
+	return !list_empty_careful(&hctx->dispatch) ||
+		sbitmap_any_bit_set(&hctx->ctx_map) ||
 			blk_mq_sched_has_work(hctx);
 }
 
@@ -125,7 +126,8 @@ void blk_freeze_queue_start(struct request_queue *q)
 	freeze_depth = atomic_inc_return(&q->mq_freeze_depth);
 	if (freeze_depth == 1) {
 		percpu_ref_kill(&q->q_usage_counter);
-		blk_mq_run_hw_queues(q, false);
+		if (q->mq_ops)
+			blk_mq_run_hw_queues(q, false);
 	}
 }
 EXPORT_SYMBOL_GPL(blk_freeze_queue_start);
@@ -255,13 +257,6 @@ void blk_mq_wake_waiters(struct request_queue *q)
 	queue_for_each_hw_ctx(q, hctx, i)
 		if (blk_mq_hw_queue_mapped(hctx))
 			blk_mq_tag_wakeup_all(hctx->tags, true);
-
-	/*
-	 * If we are called because the queue has now been marked as
-	 * dying, we need to ensure that processes currently waiting on
-	 * the queue are notified as well.
-	 */
-	wake_up_all(&q->mq_freeze_wq);
 }
 
 bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
@@ -296,6 +291,8 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data,
 	rq->q = data->q;
 	rq->mq_ctx = data->ctx;
 	rq->cmd_flags = op;
+	if (data->flags & BLK_MQ_REQ_PREEMPT)
+		rq->rq_flags |= RQF_PREEMPT;
 	if (blk_queue_io_stat(data->q))
 		rq->rq_flags |= RQF_IO_STAT;
 	/* do not touch atomic flags, it needs atomic ops against the timer */
@@ -336,12 +333,14 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 	struct elevator_queue *e = q->elevator;
 	struct request *rq;
 	unsigned int tag;
-	struct blk_mq_ctx *local_ctx = NULL;
+	bool put_ctx_on_error = false;
 
 	blk_queue_enter_live(q);
 	data->q = q;
-	if (likely(!data->ctx))
-		data->ctx = local_ctx = blk_mq_get_ctx(q);
+	if (likely(!data->ctx)) {
+		data->ctx = blk_mq_get_ctx(q);
+		put_ctx_on_error = true;
+	}
 	if (likely(!data->hctx))
 		data->hctx = blk_mq_map_queue(q, data->ctx->cpu);
 	if (op & REQ_NOWAIT)
@@ -360,8 +359,8 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 
 	tag = blk_mq_get_tag(data);
 	if (tag == BLK_MQ_TAG_FAIL) {
-		if (local_ctx) {
-			blk_mq_put_ctx(local_ctx);
+		if (put_ctx_on_error) {
+			blk_mq_put_ctx(data->ctx);
 			data->ctx = NULL;
 		}
 		blk_queue_exit(q);
@@ -384,13 +383,13 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 }
 
 struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
-		unsigned int flags)
+		blk_mq_req_flags_t flags)
 {
 	struct blk_mq_alloc_data alloc_data = { .flags = flags };
 	struct request *rq;
 	int ret;
 
-	ret = blk_queue_enter(q, flags & BLK_MQ_REQ_NOWAIT);
+	ret = blk_queue_enter(q, flags);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -410,7 +409,7 @@ struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op,
 EXPORT_SYMBOL(blk_mq_alloc_request);
 
 struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
-		unsigned int op, unsigned int flags, unsigned int hctx_idx)
+	unsigned int op, blk_mq_req_flags_t flags, unsigned int hctx_idx)
 {
 	struct blk_mq_alloc_data alloc_data = { .flags = flags };
 	struct request *rq;
@@ -429,7 +428,7 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
 	if (hctx_idx >= q->nr_hw_queues)
 		return ERR_PTR(-EIO);
 
-	ret = blk_queue_enter(q, true);
+	ret = blk_queue_enter(q, flags);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -476,8 +475,14 @@ void blk_mq_free_request(struct request *rq)
 	if (rq->rq_flags & RQF_MQ_INFLIGHT)
 		atomic_dec(&hctx->nr_active);
 
+	if (unlikely(laptop_mode && !blk_rq_is_passthrough(rq)))
+		laptop_io_completion(q->backing_dev_info);
+
 	wbt_done(q->rq_wb, &rq->issue_stat);
 
+	if (blk_rq_rl(rq))
+		blk_put_rl(blk_rq_rl(rq));
+
 	clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
 	clear_bit(REQ_ATOM_POLL_SLEPT, &rq->atomic_flags);
 	if (rq->tag != -1)
@@ -593,22 +598,32 @@ void blk_mq_start_request(struct request *rq)
 
 	blk_add_timer(rq);
 
-	/*
-	 * Ensure that ->deadline is visible before set the started
-	 * flag and clear the completed flag.
-	 */
-	smp_mb__before_atomic();
+	WARN_ON_ONCE(test_bit(REQ_ATOM_STARTED, &rq->atomic_flags));
 
 	/*
 	 * Mark us as started and clear complete. Complete might have been
 	 * set if requeue raced with timeout, which then marked it as
 	 * complete. So be sure to clear complete again when we start
 	 * the request, otherwise we'll ignore the completion event.
+	 *
+	 * Ensure that ->deadline is visible before we set STARTED, such that
+	 * blk_mq_check_expired() is guaranteed to observe our ->deadline when
+	 * it observes STARTED.
 	 */
-	if (!test_bit(REQ_ATOM_STARTED, &rq->atomic_flags))
-		set_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
-	if (test_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags))
+	smp_wmb();
+	set_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
+	if (test_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags)) {
+		/*
+		 * Coherence order guarantees these consecutive stores to a
+		 * single variable propagate in the specified order. Thus the
+		 * clear_bit() is ordered _after_ the set bit. See
+		 * blk_mq_check_expired().
+		 *
+		 * (the bits must be part of the same byte for this to be
+		 * true).
+		 */
 		clear_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags);
+	}
 
 	if (q->dma_drain_size && blk_rq_bytes(rq)) {
 		/*
@@ -634,6 +649,8 @@ static void __blk_mq_requeue_request(struct request *rq)
 {
 	struct request_queue *q = rq->q;
 
+	blk_mq_put_driver_tag(rq);
+
 	trace_block_rq_requeue(q, rq);
 	wbt_requeue(q->rq_wb, &rq->issue_stat);
 	blk_mq_sched_requeue_request(rq);
@@ -690,7 +707,7 @@ void blk_mq_add_to_requeue_list(struct request *rq, bool at_head,
 
 	/*
 	 * We abuse this flag that is otherwise used by the I/O scheduler to
-	 * request head insertation from the workqueue.
+	 * request head insertion from the workqueue.
 	 */
 	BUG_ON(rq->rq_flags & RQF_SOFTBARRIER);
 
@@ -778,11 +795,20 @@ static void blk_mq_check_expired(struct blk_mq_hw_ctx *hctx,
 		struct request *rq, void *priv, bool reserved)
 {
 	struct blk_mq_timeout_data *data = priv;
+	unsigned long deadline;
 
 	if (!test_bit(REQ_ATOM_STARTED, &rq->atomic_flags))
 		return;
 
 	/*
+	 * Ensures that if we see STARTED we must also see our
+	 * up-to-date deadline, see blk_mq_start_request().
+	 */
+	smp_rmb();
+
+	deadline = READ_ONCE(rq->deadline);
+
+	/*
 	 * The rq being checked may have been freed and reallocated
 	 * out already here, we avoid this race by checking rq->deadline
 	 * and REQ_ATOM_COMPLETE flag together:
@@ -795,11 +821,20 @@ static void blk_mq_check_expired(struct blk_mq_hw_ctx *hctx,
 	 *   and clearing the flag in blk_mq_start_request(), so
 	 *   this rq won't be timed out too.
 	 */
-	if (time_after_eq(jiffies, rq->deadline)) {
-		if (!blk_mark_rq_complete(rq))
+	if (time_after_eq(jiffies, deadline)) {
+		if (!blk_mark_rq_complete(rq)) {
+			/*
+			 * Again coherence order ensures that consecutive reads
+			 * from the same variable must be in that order. This
+			 * ensures that if we see COMPLETE clear, we must then
+			 * see STARTED set and we'll ignore this timeout.
+			 *
+			 * (There's also the MB implied by the test_and_clear())
+			 */
 			blk_mq_rq_timed_out(rq, reserved);
-	} else if (!data->next_set || time_after(data->next, rq->deadline)) {
-		data->next = rq->deadline;
+		}
+	} else if (!data->next_set || time_after(data->next, deadline)) {
+		data->next = deadline;
 		data->next_set = 1;
 	}
 }
@@ -880,6 +915,45 @@ void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list)
 }
 EXPORT_SYMBOL_GPL(blk_mq_flush_busy_ctxs);
 
+struct dispatch_rq_data {
+	struct blk_mq_hw_ctx *hctx;
+	struct request *rq;
+};
+
+static bool dispatch_rq_from_ctx(struct sbitmap *sb, unsigned int bitnr,
+		void *data)
+{
+	struct dispatch_rq_data *dispatch_data = data;
+	struct blk_mq_hw_ctx *hctx = dispatch_data->hctx;
+	struct blk_mq_ctx *ctx = hctx->ctxs[bitnr];
+
+	spin_lock(&ctx->lock);
+	if (unlikely(!list_empty(&ctx->rq_list))) {
+		dispatch_data->rq = list_entry_rq(ctx->rq_list.next);
+		list_del_init(&dispatch_data->rq->queuelist);
+		if (list_empty(&ctx->rq_list))
+			sbitmap_clear_bit(sb, bitnr);
+	}
+	spin_unlock(&ctx->lock);
+
+	return !dispatch_data->rq;
+}
+
+struct request *blk_mq_dequeue_from_ctx(struct blk_mq_hw_ctx *hctx,
+					struct blk_mq_ctx *start)
+{
+	unsigned off = start ? start->index_hw : 0;
+	struct dispatch_rq_data data = {
+		.hctx = hctx,
+		.rq   = NULL,
+	};
+
+	__sbitmap_for_each_set(&hctx->ctx_map, off,
+			       dispatch_rq_from_ctx, &data);
+
+	return data.rq;
+}
+
 static inline unsigned int queued_to_index(unsigned int queued)
 {
 	if (!queued)
@@ -920,109 +994,95 @@ done:
 	return rq->tag != -1;
 }
 
-static void __blk_mq_put_driver_tag(struct blk_mq_hw_ctx *hctx,
-				    struct request *rq)
-{
-	blk_mq_put_tag(hctx, hctx->tags, rq->mq_ctx, rq->tag);
-	rq->tag = -1;
-
-	if (rq->rq_flags & RQF_MQ_INFLIGHT) {
-		rq->rq_flags &= ~RQF_MQ_INFLIGHT;
-		atomic_dec(&hctx->nr_active);
-	}
-}
-
-static void blk_mq_put_driver_tag_hctx(struct blk_mq_hw_ctx *hctx,
-				       struct request *rq)
-{
-	if (rq->tag == -1 || rq->internal_tag == -1)
-		return;
-
-	__blk_mq_put_driver_tag(hctx, rq);
-}
-
-static void blk_mq_put_driver_tag(struct request *rq)
+static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode,
+				int flags, void *key)
 {
 	struct blk_mq_hw_ctx *hctx;
 
-	if (rq->tag == -1 || rq->internal_tag == -1)
-		return;
+	hctx = container_of(wait, struct blk_mq_hw_ctx, dispatch_wait);
 
-	hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu);
-	__blk_mq_put_driver_tag(hctx, rq);
+	list_del_init(&wait->entry);
+	blk_mq_run_hw_queue(hctx, true);
+	return 1;
 }
 
 /*
- * If we fail getting a driver tag because all the driver tags are already
- * assigned and on the dispatch list, BUT the first entry does not have a
- * tag, then we could deadlock. For that case, move entries with assigned
- * driver tags to the front, leaving the set of tagged requests in the
- * same order, and the untagged set in the same order.
+ * Mark us waiting for a tag. For shared tags, this involves hooking us into
+ * the tag wakeups. For non-shared tags, we can simply mark us nedeing a
+ * restart. For both caes, take care to check the condition again after
+ * marking us as waiting.
  */
-static bool reorder_tags_to_front(struct list_head *list)
-{
-	struct request *rq, *tmp, *first = NULL;
-
-	list_for_each_entry_safe_reverse(rq, tmp, list, queuelist) {
-		if (rq == first)
-			break;
-		if (rq->tag != -1) {
-			list_move(&rq->queuelist, list);
-			if (!first)
-				first = rq;
-		}
-	}
-
-	return first != NULL;
-}
-
-static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode, int flags,
-				void *key)
+static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx,
+				 struct request *rq)
 {
-	struct blk_mq_hw_ctx *hctx;
+	struct blk_mq_hw_ctx *this_hctx = *hctx;
+	bool shared_tags = (this_hctx->flags & BLK_MQ_F_TAG_SHARED) != 0;
+	struct sbq_wait_state *ws;
+	wait_queue_entry_t *wait;
+	bool ret;
 
-	hctx = container_of(wait, struct blk_mq_hw_ctx, dispatch_wait);
+	if (!shared_tags) {
+		if (!test_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state))
+			set_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state);
+	} else {
+		wait = &this_hctx->dispatch_wait;
+		if (!list_empty_careful(&wait->entry))
+			return false;
 
-	list_del(&wait->entry);
-	clear_bit_unlock(BLK_MQ_S_TAG_WAITING, &hctx->state);
-	blk_mq_run_hw_queue(hctx, true);
-	return 1;
-}
+		spin_lock(&this_hctx->lock);
+		if (!list_empty(&wait->entry)) {
+			spin_unlock(&this_hctx->lock);
+			return false;
+		}
 
-static bool blk_mq_dispatch_wait_add(struct blk_mq_hw_ctx *hctx)
-{
-	struct sbq_wait_state *ws;
+		ws = bt_wait_ptr(&this_hctx->tags->bitmap_tags, this_hctx);
+		add_wait_queue(&ws->wait, wait);
+	}
 
 	/*
-	 * The TAG_WAITING bit serves as a lock protecting hctx->dispatch_wait.
-	 * The thread which wins the race to grab this bit adds the hardware
-	 * queue to the wait queue.
+	 * It's possible that a tag was freed in the window between the
+	 * allocation failure and adding the hardware queue to the wait
+	 * queue.
 	 */
-	if (test_bit(BLK_MQ_S_TAG_WAITING, &hctx->state) ||
-	    test_and_set_bit_lock(BLK_MQ_S_TAG_WAITING, &hctx->state))
-		return false;
+	ret = blk_mq_get_driver_tag(rq, hctx, false);
 
-	init_waitqueue_func_entry(&hctx->dispatch_wait, blk_mq_dispatch_wake);
-	ws = bt_wait_ptr(&hctx->tags->bitmap_tags, hctx);
+	if (!shared_tags) {
+		/*
+		 * Don't clear RESTART here, someone else could have set it.
+		 * At most this will cost an extra queue run.
+		 */
+		return ret;
+	} else {
+		if (!ret) {
+			spin_unlock(&this_hctx->lock);
+			return false;
+		}
 
-	/*
-	 * As soon as this returns, it's no longer safe to fiddle with
-	 * hctx->dispatch_wait, since a completion can wake up the wait queue
-	 * and unlock the bit.
-	 */
-	add_wait_queue(&ws->wait, &hctx->dispatch_wait);
-	return true;
+		/*
+		 * We got a tag, remove ourselves from the wait queue to ensure
+		 * someone else gets the wakeup.
+		 */
+		spin_lock_irq(&ws->wait.lock);
+		list_del_init(&wait->entry);
+		spin_unlock_irq(&ws->wait.lock);
+		spin_unlock(&this_hctx->lock);
+		return true;
+	}
 }
 
-bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list)
+bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list,
+			     bool got_budget)
 {
 	struct blk_mq_hw_ctx *hctx;
-	struct request *rq;
+	struct request *rq, *nxt;
+	bool no_tag = false;
 	int errors, queued;
 
 	if (list_empty(list))
 		return false;
 
+	WARN_ON(!list_is_singular(list) && got_budget);
+
 	/*
 	 * Now process all the entries, sending them to the driver.
 	 */
@@ -1033,23 +1093,29 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list)
 
 		rq = list_first_entry(list, struct request, queuelist);
 		if (!blk_mq_get_driver_tag(rq, &hctx, false)) {
-			if (!queued && reorder_tags_to_front(list))
-				continue;
-
 			/*
 			 * The initial allocation attempt failed, so we need to
-			 * rerun the hardware queue when a tag is freed.
+			 * rerun the hardware queue when a tag is freed. The
+			 * waitqueue takes care of that. If the queue is run
+			 * before we add this entry back on the dispatch list,
+			 * we'll re-run it below.
 			 */
-			if (!blk_mq_dispatch_wait_add(hctx))
+			if (!blk_mq_mark_tag_wait(&hctx, rq)) {
+				if (got_budget)
+					blk_mq_put_dispatch_budget(hctx);
+				/*
+				 * For non-shared tags, the RESTART check
+				 * will suffice.
+				 */
+				if (hctx->flags & BLK_MQ_F_TAG_SHARED)
+					no_tag = true;
 				break;
+			}
+		}
 
-			/*
-			 * It's possible that a tag was freed in the window
-			 * between the allocation failure and adding the
-			 * hardware queue to the wait queue.
-			 */
-			if (!blk_mq_get_driver_tag(rq, &hctx, false))
-				break;
+		if (!got_budget && !blk_mq_get_dispatch_budget(hctx)) {
+			blk_mq_put_driver_tag(rq);
+			break;
 		}
 
 		list_del_init(&rq->queuelist);
@@ -1063,15 +1129,21 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list)
 		if (list_empty(list))
 			bd.last = true;
 		else {
-			struct request *nxt;
-
 			nxt = list_first_entry(list, struct request, queuelist);
 			bd.last = !blk_mq_get_driver_tag(nxt, NULL, false);
 		}
 
 		ret = q->mq_ops->queue_rq(hctx, &bd);
 		if (ret == BLK_STS_RESOURCE) {
-			blk_mq_put_driver_tag_hctx(hctx, rq);
+			/*
+			 * If an I/O scheduler has been configured and we got a
+			 * driver tag for the next request already, free it
+			 * again.
+			 */
+			if (!list_empty(list)) {
+				nxt = list_first_entry(list, struct request, queuelist);
+				blk_mq_put_driver_tag(nxt);
+			}
 			list_add(&rq->queuelist, list);
 			__blk_mq_requeue_request(rq);
 			break;
@@ -1093,13 +1165,6 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list)
 	 * that is where we will continue on next queue run.
 	 */
 	if (!list_empty(list)) {
-		/*
-		 * If an I/O scheduler has been configured and we got a driver
-		 * tag for the next request already, free it again.
-		 */
-		rq = list_first_entry(list, struct request, queuelist);
-		blk_mq_put_driver_tag(rq);
-
 		spin_lock(&hctx->lock);
 		list_splice_init(list, &hctx->dispatch);
 		spin_unlock(&hctx->lock);
@@ -1109,10 +1174,10 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list)
 		 * it is no longer set that means that it was cleared by another
 		 * thread and hence that a queue rerun is needed.
 		 *
-		 * If TAG_WAITING is set that means that an I/O scheduler has
-		 * been configured and another thread is waiting for a driver
-		 * tag. To guarantee fairness, do not rerun this hardware queue
-		 * but let the other thread grab the driver tag.
+		 * If 'no_tag' is set, that means that we failed getting
+		 * a driver tag with an I/O scheduler attached. If our dispatch
+		 * waitqueue is no longer active, ensure that we run the queue
+		 * AFTER adding our entries back to the list.
 		 *
 		 * If no I/O scheduler has been configured it is possible that
 		 * the hardware queue got stopped and restarted before requests
@@ -1124,8 +1189,8 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list)
 		 *   returning BLK_STS_RESOURCE. Two exceptions are scsi-mq
 		 *   and dm-rq.
 		 */
-		if (!blk_mq_sched_needs_restart(hctx) &&
-		    !test_bit(BLK_MQ_S_TAG_WAITING, &hctx->state))
+		if (!blk_mq_sched_needs_restart(hctx) ||
+		    (no_tag && list_empty_careful(&hctx->dispatch_wait.entry)))
 			blk_mq_run_hw_queue(hctx, true);
 	}
 
@@ -1218,9 +1283,14 @@ void blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs)
 }
 EXPORT_SYMBOL(blk_mq_delay_run_hw_queue);
 
-void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
+bool blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
 {
-	__blk_mq_delay_run_hw_queue(hctx, async, 0);
+	if (blk_mq_hctx_has_pending(hctx)) {
+		__blk_mq_delay_run_hw_queue(hctx, async, 0);
+		return true;
+	}
+
+	return false;
 }
 EXPORT_SYMBOL(blk_mq_run_hw_queue);
 
@@ -1230,8 +1300,7 @@ void blk_mq_run_hw_queues(struct request_queue *q, bool async)
 	int i;
 
 	queue_for_each_hw_ctx(q, hctx, i) {
-		if (!blk_mq_hctx_has_pending(hctx) ||
-		    blk_mq_hctx_stopped(hctx))
+		if (blk_mq_hctx_stopped(hctx))
 			continue;
 
 		blk_mq_run_hw_queue(hctx, async);
@@ -1405,7 +1474,7 @@ void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
  * Should only be used carefully, when the caller knows we want to
  * bypass a potential IO scheduler on the target device.
  */
-void blk_mq_request_bypass_insert(struct request *rq)
+void blk_mq_request_bypass_insert(struct request *rq, bool run_queue)
 {
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 	struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(rq->q, ctx->cpu);
@@ -1414,7 +1483,8 @@ void blk_mq_request_bypass_insert(struct request *rq)
 	list_add_tail(&rq->queuelist, &hctx->dispatch);
 	spin_unlock(&hctx->lock);
 
-	blk_mq_run_hw_queue(hctx, false);
+	if (run_queue)
+		blk_mq_run_hw_queue(hctx, false);
 }
 
 void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx,
@@ -1501,13 +1571,9 @@ static void blk_mq_bio_to_request(struct request *rq, struct bio *bio)
 {
 	blk_init_request_from_bio(rq, bio);
 
-	blk_account_io_start(rq, true);
-}
+	blk_rq_set_rl(rq, blk_get_rl(rq->q, bio));
 
-static inline bool hctx_allow_merges(struct blk_mq_hw_ctx *hctx)
-{
-	return (hctx->flags & BLK_MQ_F_SHOULD_MERGE) &&
-		!blk_queue_nomerges(hctx->queue);
+	blk_account_io_start(rq, true);
 }
 
 static inline void blk_mq_queue_io(struct blk_mq_hw_ctx *hctx,
@@ -1552,6 +1618,11 @@ static void __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 	if (!blk_mq_get_driver_tag(rq, NULL, false))
 		goto insert;
 
+	if (!blk_mq_get_dispatch_budget(hctx)) {
+		blk_mq_put_driver_tag(rq);
+		goto insert;
+	}
+
 	new_cookie = request_to_qc_t(hctx, rq);
 
 	/*
@@ -1641,13 +1712,10 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
 	if (unlikely(is_flush_fua)) {
 		blk_mq_put_ctx(data.ctx);
 		blk_mq_bio_to_request(rq, bio);
-		if (q->elevator) {
-			blk_mq_sched_insert_request(rq, false, true, true,
-					true);
-		} else {
-			blk_insert_flush(rq);
-			blk_mq_run_hw_queue(data.hctx, true);
-		}
+
+		/* bypass scheduler for flush rq */
+		blk_insert_flush(rq);
+		blk_mq_run_hw_queue(data.hctx, true);
 	} else if (plug && q->nr_hw_queues == 1) {
 		struct request *last = NULL;
 
@@ -1990,6 +2058,9 @@ static int blk_mq_init_hctx(struct request_queue *q,
 
 	hctx->nr_ctx = 0;
 
+	init_waitqueue_func_entry(&hctx->dispatch_wait, blk_mq_dispatch_wake);
+	INIT_LIST_HEAD(&hctx->dispatch_wait.entry);
+
 	if (set->ops->init_hctx &&
 	    set->ops->init_hctx(hctx, set->driver_data, hctx_idx))
 		goto free_bitmap;
@@ -2229,8 +2300,11 @@ static void blk_mq_add_queue_tag_set(struct blk_mq_tag_set *set,
 
 	mutex_lock(&set->tag_list_lock);
 
-	/* Check to see if we're transitioning to shared (from 1 to 2 queues). */
-	if (!list_empty(&set->tag_list) && !(set->flags & BLK_MQ_F_TAG_SHARED)) {
+	/*
+	 * Check to see if we're transitioning to shared (from 1 to 2 queues).
+	 */
+	if (!list_empty(&set->tag_list) &&
+	    !(set->flags & BLK_MQ_F_TAG_SHARED)) {
 		set->flags |= BLK_MQ_F_TAG_SHARED;
 		/* update existing queue */
 		blk_mq_update_tag_set_depth(set, true);
@@ -2404,6 +2478,8 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	spin_lock_init(&q->requeue_lock);
 
 	blk_queue_make_request(q, blk_mq_make_request);
+	if (q->mq_ops->poll)
+		q->poll_fn = blk_mq_poll;
 
 	/*
 	 * Do this after blk_queue_make_request() overrides it...
@@ -2460,10 +2536,9 @@ static void blk_mq_queue_reinit(struct request_queue *q)
 
 	/*
 	 * redo blk_mq_init_cpu_queues and blk_mq_init_hw_queues. FIXME: maybe
-	 * we should change hctx numa_node according to new topology (this
-	 * involves free and re-allocate memory, worthy doing?)
+	 * we should change hctx numa_node according to the new topology (this
+	 * involves freeing and re-allocating memory, worth doing?)
 	 */
-
 	blk_mq_map_swqueue(q);
 
 	blk_mq_sysfs_register(q);
@@ -2552,6 +2627,9 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
 	if (!set->ops->queue_rq)
 		return -EINVAL;
 
+	if (!set->ops->get_budget ^ !set->ops->put_budget)
+		return -EINVAL;
+
 	if (set->queue_depth > BLK_MQ_MAX_DEPTH) {
 		pr_info("blk-mq: reduced tag depth to %u\n",
 			BLK_MQ_MAX_DEPTH);
@@ -2642,8 +2720,7 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
 		 * queue depth. This is similar to what the old code would do.
 		 */
 		if (!hctx->sched_tags) {
-			ret = blk_mq_tag_update_depth(hctx, &hctx->tags,
-							min(nr, set->queue_depth),
+			ret = blk_mq_tag_update_depth(hctx, &hctx->tags, nr,
 							false);
 		} else {
 			ret = blk_mq_tag_update_depth(hctx, &hctx->sched_tags,
@@ -2863,20 +2940,14 @@ static bool __blk_mq_poll(struct blk_mq_hw_ctx *hctx, struct request *rq)
 	return false;
 }
 
-bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie)
+static bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie)
 {
 	struct blk_mq_hw_ctx *hctx;
-	struct blk_plug *plug;
 	struct request *rq;
 
-	if (!q->mq_ops || !q->mq_ops->poll || !blk_qc_t_valid(cookie) ||
-	    !test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
+	if (!test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
 		return false;
 
-	plug = current->plug;
-	if (plug)
-		blk_flush_plug_list(plug, false);
-
 	hctx = q->queue_hw_ctx[blk_qc_t_to_queue_num(cookie)];
 	if (!blk_qc_t_is_internal(cookie))
 		rq = blk_mq_tag_to_rq(hctx->tags, blk_qc_t_to_tag(cookie));
@@ -2894,10 +2965,15 @@ bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie)
 
 	return __blk_mq_poll(hctx, rq);
 }
-EXPORT_SYMBOL_GPL(blk_mq_poll);
 
 static int __init blk_mq_init(void)
 {
+	/*
+	 * See comment in block/blk.h rq_atomic_flags enum
+	 */
+	BUILD_BUG_ON((REQ_ATOM_STARTED / BITS_PER_BYTE) !=
+			(REQ_ATOM_COMPLETE / BITS_PER_BYTE));
+
 	cpuhp_setup_state_multi(CPUHP_BLK_MQ_DEAD, "block/mq:dead", NULL,
 				blk_mq_hctx_notify_dead);
 	return 0;
diff --git a/block/blk-mq.h b/block/blk-mq.h
index 4933af9d61f7..6c7c3ff5bf62 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -3,6 +3,7 @@
 #define INT_BLK_MQ_H
 
 #include "blk-stat.h"
+#include "blk-mq-tag.h"
 
 struct blk_mq_tag_set;
 
@@ -26,16 +27,16 @@ struct blk_mq_ctx {
 	struct kobject		kobj;
 } ____cacheline_aligned_in_smp;
 
-void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async);
 void blk_mq_freeze_queue(struct request_queue *q);
 void blk_mq_free_queue(struct request_queue *q);
 int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr);
 void blk_mq_wake_waiters(struct request_queue *q);
-bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *);
+bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *, bool);
 void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list);
-bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx);
 bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx,
 				bool wait);
+struct request *blk_mq_dequeue_from_ctx(struct blk_mq_hw_ctx *hctx,
+					struct blk_mq_ctx *start);
 
 /*
  * Internal helpers for allocating/freeing the request map
@@ -55,7 +56,7 @@ int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags,
  */
 void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 				bool at_head);
-void blk_mq_request_bypass_insert(struct request *rq);
+void blk_mq_request_bypass_insert(struct request *rq, bool run_queue);
 void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx,
 				struct list_head *list);
 
@@ -109,7 +110,7 @@ static inline void blk_mq_put_ctx(struct blk_mq_ctx *ctx)
 struct blk_mq_alloc_data {
 	/* input parameter */
 	struct request_queue *q;
-	unsigned int flags;
+	blk_mq_req_flags_t flags;
 	unsigned int shallow_depth;
 
 	/* input & output parameter */
@@ -138,4 +139,53 @@ static inline bool blk_mq_hw_queue_mapped(struct blk_mq_hw_ctx *hctx)
 void blk_mq_in_flight(struct request_queue *q, struct hd_struct *part,
 			unsigned int inflight[2]);
 
+static inline void blk_mq_put_dispatch_budget(struct blk_mq_hw_ctx *hctx)
+{
+	struct request_queue *q = hctx->queue;
+
+	if (q->mq_ops->put_budget)
+		q->mq_ops->put_budget(hctx);
+}
+
+static inline bool blk_mq_get_dispatch_budget(struct blk_mq_hw_ctx *hctx)
+{
+	struct request_queue *q = hctx->queue;
+
+	if (q->mq_ops->get_budget)
+		return q->mq_ops->get_budget(hctx);
+	return true;
+}
+
+static inline void __blk_mq_put_driver_tag(struct blk_mq_hw_ctx *hctx,
+					   struct request *rq)
+{
+	blk_mq_put_tag(hctx, hctx->tags, rq->mq_ctx, rq->tag);
+	rq->tag = -1;
+
+	if (rq->rq_flags & RQF_MQ_INFLIGHT) {
+		rq->rq_flags &= ~RQF_MQ_INFLIGHT;
+		atomic_dec(&hctx->nr_active);
+	}
+}
+
+static inline void blk_mq_put_driver_tag_hctx(struct blk_mq_hw_ctx *hctx,
+				       struct request *rq)
+{
+	if (rq->tag == -1 || rq->internal_tag == -1)
+		return;
+
+	__blk_mq_put_driver_tag(hctx, rq);
+}
+
+static inline void blk_mq_put_driver_tag(struct request *rq)
+{
+	struct blk_mq_hw_ctx *hctx;
+
+	if (rq->tag == -1 || rq->internal_tag == -1)
+		return;
+
+	hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu);
+	__blk_mq_put_driver_tag(hctx, rq);
+}
+
 #endif
diff --git a/block/blk-settings.c b/block/blk-settings.c
index 8559e9563c52..48ebe6be07b7 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -157,7 +157,7 @@ EXPORT_SYMBOL(blk_set_stacking_limits);
  * Caveat:
  *    The driver that does this *must* be able to deal appropriately
  *    with buffers in "highmemory". This can be accomplished by either calling
- *    __bio_kmap_atomic() to get a temporary kernel mapping, or by calling
+ *    kmap_atomic() to get a temporary kernel mapping, or by calling
  *    blk_queue_bounce() to create a buffer in normal memory.
  **/
 void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn)
diff --git a/block/blk-stat.c b/block/blk-stat.c
index c52356d90fe3..3a2f3c96f367 100644
--- a/block/blk-stat.c
+++ b/block/blk-stat.c
@@ -11,8 +11,6 @@
 #include "blk-mq.h"
 #include "blk.h"
 
-#define BLK_RQ_STAT_BATCH	64
-
 struct blk_queue_stats {
 	struct list_head callbacks;
 	spinlock_t lock;
@@ -23,45 +21,21 @@ static void blk_stat_init(struct blk_rq_stat *stat)
 {
 	stat->min = -1ULL;
 	stat->max = stat->nr_samples = stat->mean = 0;
-	stat->batch = stat->nr_batch = 0;
-}
-
-static void blk_stat_flush_batch(struct blk_rq_stat *stat)
-{
-	const s32 nr_batch = READ_ONCE(stat->nr_batch);
-	const s32 nr_samples = READ_ONCE(stat->nr_samples);
-
-	if (!nr_batch)
-		return;
-	if (!nr_samples)
-		stat->mean = div64_s64(stat->batch, nr_batch);
-	else {
-		stat->mean = div64_s64((stat->mean * nr_samples) +
-					stat->batch,
-					nr_batch + nr_samples);
-	}
-
-	stat->nr_samples += nr_batch;
-	stat->nr_batch = stat->batch = 0;
+	stat->batch = 0;
 }
 
+/* src is a per-cpu stat, mean isn't initialized */
 static void blk_stat_sum(struct blk_rq_stat *dst, struct blk_rq_stat *src)
 {
-	blk_stat_flush_batch(src);
-
 	if (!src->nr_samples)
 		return;
 
 	dst->min = min(dst->min, src->min);
 	dst->max = max(dst->max, src->max);
 
-	if (!dst->nr_samples)
-		dst->mean = src->mean;
-	else {
-		dst->mean = div64_s64((src->mean * src->nr_samples) +
-					(dst->mean * dst->nr_samples),
-					dst->nr_samples + src->nr_samples);
-	}
+	dst->mean = div_u64(src->batch + dst->mean * dst->nr_samples,
+				dst->nr_samples + src->nr_samples);
+
 	dst->nr_samples += src->nr_samples;
 }
 
@@ -69,13 +43,8 @@ static void __blk_stat_add(struct blk_rq_stat *stat, u64 value)
 {
 	stat->min = min(stat->min, value);
 	stat->max = max(stat->max, value);
-
-	if (stat->batch + value < stat->batch ||
-	    stat->nr_batch + 1 == BLK_RQ_STAT_BATCH)
-		blk_stat_flush_batch(stat);
-
 	stat->batch += value;
-	stat->nr_batch++;
+	stat->nr_samples++;
 }
 
 void blk_stat_add(struct request *rq)
@@ -84,7 +53,7 @@ void blk_stat_add(struct request *rq)
 	struct blk_stat_callback *cb;
 	struct blk_rq_stat *stat;
 	int bucket;
-	s64 now, value;
+	u64 now, value;
 
 	now = __blk_stat_time(ktime_to_ns(ktime_get()));
 	if (now < blk_stat_time(&rq->issue_stat))
diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 8631763866c6..96ad32623427 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -2113,8 +2113,12 @@ static inline void throtl_update_latency_buckets(struct throtl_data *td)
 static void blk_throtl_assoc_bio(struct throtl_grp *tg, struct bio *bio)
 {
 #ifdef CONFIG_BLK_DEV_THROTTLING_LOW
-	if (bio->bi_css)
+	if (bio->bi_css) {
+		if (bio->bi_cg_private)
+			blkg_put(tg_to_blkg(bio->bi_cg_private));
 		bio->bi_cg_private = tg;
+		blkg_get(tg_to_blkg(tg));
+	}
 	blk_stat_set_issue(&bio->bi_issue_stat, bio_sectors(bio));
 #endif
 }
@@ -2284,8 +2288,10 @@ void blk_throtl_bio_endio(struct bio *bio)
 
 	start_time = blk_stat_time(&bio->bi_issue_stat) >> 10;
 	finish_time = __blk_stat_time(finish_time_ns) >> 10;
-	if (!start_time || finish_time <= start_time)
+	if (!start_time || finish_time <= start_time) {
+		blkg_put(tg_to_blkg(tg));
 		return;
+	}
 
 	lat = finish_time - start_time;
 	/* this is only for bio based driver */
@@ -2315,6 +2321,8 @@ void blk_throtl_bio_endio(struct bio *bio)
 		tg->bio_cnt /= 2;
 		tg->bad_bio_cnt /= 2;
 	}
+
+	blkg_put(tg_to_blkg(tg));
 }
 #endif
 
diff --git a/block/blk-timeout.c b/block/blk-timeout.c
index 17ec83bb0900..764ecf9aeb30 100644
--- a/block/blk-timeout.c
+++ b/block/blk-timeout.c
@@ -134,8 +134,6 @@ void blk_timeout_work(struct work_struct *work)
 	struct request *rq, *tmp;
 	int next_set = 0;
 
-	if (blk_queue_enter(q, true))
-		return;
 	spin_lock_irqsave(q->queue_lock, flags);
 
 	list_for_each_entry_safe(rq, tmp, &q->timeout_list, timeout_list)
@@ -145,7 +143,6 @@ void blk_timeout_work(struct work_struct *work)
 		mod_timer(&q->timeout, round_jiffies_up(next));
 
 	spin_unlock_irqrestore(q->queue_lock, flags);
-	blk_queue_exit(q);
 }
 
 /**
@@ -211,7 +208,7 @@ void blk_add_timer(struct request *req)
 	if (!req->timeout)
 		req->timeout = q->rq_timeout;
 
-	req->deadline = jiffies + req->timeout;
+	WRITE_ONCE(req->deadline, jiffies + req->timeout);
 
 	/*
 	 * Only the non-mq case needs to add the request to a protected list.
diff --git a/block/blk-wbt.c b/block/blk-wbt.c
index d822530e6aea..b252da0e4c11 100644
--- a/block/blk-wbt.c
+++ b/block/blk-wbt.c
@@ -654,7 +654,7 @@ void wbt_set_write_cache(struct rq_wb *rwb, bool write_cache_on)
 }
 
 /*
- * Disable wbt, if enabled by default. Only called from CFQ.
+ * Disable wbt, if enabled by default.
  */
 void wbt_disable_default(struct request_queue *q)
 {
diff --git a/block/blk.h b/block/blk.h
index 85be8b232b37..3f1446937aec 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -123,8 +123,15 @@ void blk_account_io_done(struct request *req);
  * Internal atomic flags for request handling
  */
 enum rq_atomic_flags {
+	/*
+	 * Keep these two bits first - not because we depend on the
+	 * value of them, but we do depend on them being in the same
+	 * byte of storage to ensure ordering on writes. Keeping them
+	 * first will achieve that nicely.
+	 */
 	REQ_ATOM_COMPLETE = 0,
 	REQ_ATOM_STARTED,
+
 	REQ_ATOM_POLL_SLEPT,
 };
 
@@ -149,45 +156,6 @@ static inline void blk_clear_rq_complete(struct request *rq)
 
 void blk_insert_flush(struct request *rq);
 
-static inline struct request *__elv_next_request(struct request_queue *q)
-{
-	struct request *rq;
-	struct blk_flush_queue *fq = blk_get_flush_queue(q, NULL);
-
-	WARN_ON_ONCE(q->mq_ops);
-
-	while (1) {
-		if (!list_empty(&q->queue_head)) {
-			rq = list_entry_rq(q->queue_head.next);
-			return rq;
-		}
-
-		/*
-		 * Flush request is running and flush request isn't queueable
-		 * in the drive, we can hold the queue till flush request is
-		 * finished. Even we don't do this, driver can't dispatch next
-		 * requests and will requeue them. And this can improve
-		 * throughput too. For example, we have request flush1, write1,
-		 * flush 2. flush1 is dispatched, then queue is hold, write1
-		 * isn't inserted to queue. After flush1 is finished, flush2
-		 * will be dispatched. Since disk cache is already clean,
-		 * flush2 will be finished very soon, so looks like flush2 is
-		 * folded to flush1.
-		 * Since the queue is hold, a flag is set to indicate the queue
-		 * should be restarted later. Please see flush_end_io() for
-		 * details.
-		 */
-		if (fq->flush_pending_idx != fq->flush_running_idx &&
-				!queue_flush_queueable(q)) {
-			fq->flush_queue_delayed = 1;
-			return NULL;
-		}
-		if (unlikely(blk_queue_bypass(q)) ||
-		    !q->elevator->type->ops.sq.elevator_dispatch_fn(q, 0))
-			return NULL;
-	}
-}
-
 static inline void elv_activate_rq(struct request_queue *q, struct request *rq)
 {
 	struct elevator_queue *e = q->elevator;
diff --git a/block/bsg.c b/block/bsg.c
index ee1335c68de7..452f94f1c5d4 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -137,7 +137,7 @@ static inline struct hlist_head *bsg_dev_idx_hash(int index)
 
 static int blk_fill_sgv4_hdr_rq(struct request_queue *q, struct request *rq,
 				struct sg_io_v4 *hdr, struct bsg_device *bd,
-				fmode_t has_write_perm)
+				fmode_t mode)
 {
 	struct scsi_request *req = scsi_req(rq);
 
@@ -152,7 +152,7 @@ static int blk_fill_sgv4_hdr_rq(struct request_queue *q, struct request *rq,
 		return -EFAULT;
 
 	if (hdr->subprotocol == BSG_SUB_PROTOCOL_SCSI_CMD) {
-		if (blk_verify_command(req->cmd, has_write_perm))
+		if (blk_verify_command(req->cmd, mode))
 			return -EPERM;
 	} else if (!capable(CAP_SYS_RAWIO))
 		return -EPERM;
@@ -206,7 +206,7 @@ bsg_validate_sgv4_hdr(struct sg_io_v4 *hdr, int *op)
  * map sg_io_v4 to a request.
  */
 static struct request *
-bsg_map_hdr(struct bsg_device *bd, struct sg_io_v4 *hdr, fmode_t has_write_perm)
+bsg_map_hdr(struct bsg_device *bd, struct sg_io_v4 *hdr, fmode_t mode)
 {
 	struct request_queue *q = bd->queue;
 	struct request *rq, *next_rq = NULL;
@@ -237,7 +237,7 @@ bsg_map_hdr(struct bsg_device *bd, struct sg_io_v4 *hdr, fmode_t has_write_perm)
 	if (IS_ERR(rq))
 		return rq;
 
-	ret = blk_fill_sgv4_hdr_rq(q, rq, hdr, bd, has_write_perm);
+	ret = blk_fill_sgv4_hdr_rq(q, rq, hdr, bd, mode);
 	if (ret)
 		goto out;
 
@@ -587,8 +587,7 @@ bsg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 }
 
 static int __bsg_write(struct bsg_device *bd, const char __user *buf,
-		       size_t count, ssize_t *bytes_written,
-		       fmode_t has_write_perm)
+		       size_t count, ssize_t *bytes_written, fmode_t mode)
 {
 	struct bsg_command *bc;
 	struct request *rq;
@@ -619,7 +618,7 @@ static int __bsg_write(struct bsg_device *bd, const char __user *buf,
 		/*
 		 * get a request, fill in the blanks, and add to request queue
 		 */
-		rq = bsg_map_hdr(bd, &bc->hdr, has_write_perm);
+		rq = bsg_map_hdr(bd, &bc->hdr, mode);
 		if (IS_ERR(rq)) {
 			ret = PTR_ERR(rq);
 			rq = NULL;
@@ -655,8 +654,7 @@ bsg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	bsg_set_block(bd, file);
 
 	bytes_written = 0;
-	ret = __bsg_write(bd, buf, count, &bytes_written,
-			  file->f_mode & FMODE_WRITE);
+	ret = __bsg_write(bd, buf, count, &bytes_written, file->f_mode);
 
 	*ppos = bytes_written;
 
@@ -915,7 +913,7 @@ static long bsg_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		if (copy_from_user(&hdr, uarg, sizeof(hdr)))
 			return -EFAULT;
 
-		rq = bsg_map_hdr(bd, &hdr, file->f_mode & FMODE_WRITE);
+		rq = bsg_map_hdr(bd, &hdr, file->f_mode);
 		if (IS_ERR(rq))
 			return PTR_ERR(rq);
 
diff --git a/block/elevator.c b/block/elevator.c
index 153926a90901..7bda083d5968 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -83,12 +83,25 @@ bool elv_bio_merge_ok(struct request *rq, struct bio *bio)
 }
 EXPORT_SYMBOL(elv_bio_merge_ok);
 
-static struct elevator_type *elevator_find(const char *name)
+static bool elevator_match(const struct elevator_type *e, const char *name)
+{
+	if (!strcmp(e->elevator_name, name))
+		return true;
+	if (e->elevator_alias && !strcmp(e->elevator_alias, name))
+		return true;
+
+	return false;
+}
+
+/*
+ * Return scheduler with name 'name' and with matching 'mq capability
+ */
+static struct elevator_type *elevator_find(const char *name, bool mq)
 {
 	struct elevator_type *e;
 
 	list_for_each_entry(e, &elv_list, list) {
-		if (!strcmp(e->elevator_name, name))
+		if (elevator_match(e, name) && (mq == e->uses_mq))
 			return e;
 	}
 
@@ -100,25 +113,25 @@ static void elevator_put(struct elevator_type *e)
 	module_put(e->elevator_owner);
 }
 
-static struct elevator_type *elevator_get(const char *name, bool try_loading)
+static struct elevator_type *elevator_get(struct request_queue *q,
+					  const char *name, bool try_loading)
 {
 	struct elevator_type *e;
 
 	spin_lock(&elv_list_lock);
 
-	e = elevator_find(name);
+	e = elevator_find(name, q->mq_ops != NULL);
 	if (!e && try_loading) {
 		spin_unlock(&elv_list_lock);
 		request_module("%s-iosched", name);
 		spin_lock(&elv_list_lock);
-		e = elevator_find(name);
+		e = elevator_find(name, q->mq_ops != NULL);
 	}
 
 	if (e && !try_module_get(e->elevator_owner))
 		e = NULL;
 
 	spin_unlock(&elv_list_lock);
-
 	return e;
 }
 
@@ -144,8 +157,12 @@ void __init load_default_elevator_module(void)
 	if (!chosen_elevator[0])
 		return;
 
+	/*
+	 * Boot parameter is deprecated, we haven't supported that for MQ.
+	 * Only look for non-mq schedulers from here.
+	 */
 	spin_lock(&elv_list_lock);
-	e = elevator_find(chosen_elevator);
+	e = elevator_find(chosen_elevator, false);
 	spin_unlock(&elv_list_lock);
 
 	if (!e)
@@ -202,7 +219,7 @@ int elevator_init(struct request_queue *q, char *name)
 	q->boundary_rq = NULL;
 
 	if (name) {
-		e = elevator_get(name, true);
+		e = elevator_get(q, name, true);
 		if (!e)
 			return -EINVAL;
 	}
@@ -214,7 +231,7 @@ int elevator_init(struct request_queue *q, char *name)
 	 * allowed from async.
 	 */
 	if (!e && !q->mq_ops && *chosen_elevator) {
-		e = elevator_get(chosen_elevator, false);
+		e = elevator_get(q, chosen_elevator, false);
 		if (!e)
 			printk(KERN_ERR "I/O scheduler %s not found\n",
 							chosen_elevator);
@@ -229,17 +246,17 @@ int elevator_init(struct request_queue *q, char *name)
 		 */
 		if (q->mq_ops) {
 			if (q->nr_hw_queues == 1)
-				e = elevator_get("mq-deadline", false);
+				e = elevator_get(q, "mq-deadline", false);
 			if (!e)
 				return 0;
 		} else
-			e = elevator_get(CONFIG_DEFAULT_IOSCHED, false);
+			e = elevator_get(q, CONFIG_DEFAULT_IOSCHED, false);
 
 		if (!e) {
 			printk(KERN_ERR
 				"Default I/O scheduler not found. " \
 				"Using noop.\n");
-			e = elevator_get("noop", false);
+			e = elevator_get(q, "noop", false);
 		}
 	}
 
@@ -905,7 +922,7 @@ int elv_register(struct elevator_type *e)
 
 	/* register, don't allow duplicate names */
 	spin_lock(&elv_list_lock);
-	if (elevator_find(e->elevator_name)) {
+	if (elevator_find(e->elevator_name, e->uses_mq)) {
 		spin_unlock(&elv_list_lock);
 		if (e->icq_cache)
 			kmem_cache_destroy(e->icq_cache);
@@ -915,9 +932,9 @@ int elv_register(struct elevator_type *e)
 	spin_unlock(&elv_list_lock);
 
 	/* print pretty message */
-	if (!strcmp(e->elevator_name, chosen_elevator) ||
+	if (elevator_match(e, chosen_elevator) ||
 			(!*chosen_elevator &&
-			 !strcmp(e->elevator_name, CONFIG_DEFAULT_IOSCHED)))
+			 elevator_match(e, CONFIG_DEFAULT_IOSCHED)))
 				def = " (default)";
 
 	printk(KERN_INFO "io scheduler %s registered%s\n", e->elevator_name,
@@ -1066,25 +1083,15 @@ static int __elevator_change(struct request_queue *q, const char *name)
 		return elevator_switch(q, NULL);
 
 	strlcpy(elevator_name, name, sizeof(elevator_name));
-	e = elevator_get(strstrip(elevator_name), true);
+	e = elevator_get(q, strstrip(elevator_name), true);
 	if (!e)
 		return -EINVAL;
 
-	if (q->elevator &&
-	    !strcmp(elevator_name, q->elevator->type->elevator_name)) {
+	if (q->elevator && elevator_match(q->elevator->type, elevator_name)) {
 		elevator_put(e);
 		return 0;
 	}
 
-	if (!e->uses_mq && q->mq_ops) {
-		elevator_put(e);
-		return -EINVAL;
-	}
-	if (e->uses_mq && !q->mq_ops) {
-		elevator_put(e);
-		return -EINVAL;
-	}
-
 	return elevator_switch(q, e);
 }
 
@@ -1116,9 +1123,10 @@ ssize_t elv_iosched_show(struct request_queue *q, char *name)
 	struct elevator_queue *e = q->elevator;
 	struct elevator_type *elv = NULL;
 	struct elevator_type *__e;
+	bool uses_mq = q->mq_ops != NULL;
 	int len = 0;
 
-	if (!blk_queue_stackable(q))
+	if (!queue_is_rq_based(q))
 		return sprintf(name, "none\n");
 
 	if (!q->elevator)
@@ -1128,7 +1136,8 @@ ssize_t elv_iosched_show(struct request_queue *q, char *name)
 
 	spin_lock(&elv_list_lock);
 	list_for_each_entry(__e, &elv_list, list) {
-		if (elv && !strcmp(elv->elevator_name, __e->elevator_name)) {
+		if (elv && elevator_match(elv, __e->elevator_name) &&
+		    (__e->uses_mq == uses_mq)) {
 			len += sprintf(name+len, "[%s] ", elv->elevator_name);
 			continue;
 		}
diff --git a/block/genhd.c b/block/genhd.c
index 630c0da6cfcf..c2223f12a805 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -588,6 +588,11 @@ static void register_disk(struct device *parent, struct gendisk *disk)
 	disk->part0.holder_dir = kobject_create_and_add("holders", &ddev->kobj);
 	disk->slave_dir = kobject_create_and_add("slaves", &ddev->kobj);
 
+	if (disk->flags & GENHD_FL_HIDDEN) {
+		dev_set_uevent_suppress(ddev, 0);
+		return;
+	}
+
 	/* No minors to use for partitions */
 	if (!disk_part_scan_enabled(disk))
 		goto exit;
@@ -616,6 +621,11 @@ exit:
 	while ((part = disk_part_iter_next(&piter)))
 		kobject_uevent(&part_to_dev(part)->kobj, KOBJ_ADD);
 	disk_part_iter_exit(&piter);
+
+	err = sysfs_create_link(&ddev->kobj,
+				&disk->queue->backing_dev_info->dev->kobj,
+				"bdi");
+	WARN_ON(err);
 }
 
 /**
@@ -630,7 +640,6 @@ exit:
  */
 void device_add_disk(struct device *parent, struct gendisk *disk)
 {
-	struct backing_dev_info *bdi;
 	dev_t devt;
 	int retval;
 
@@ -639,7 +648,8 @@ void device_add_disk(struct device *parent, struct gendisk *disk)
 	 * parameters make sense.
 	 */
 	WARN_ON(disk->minors && !(disk->major || disk->first_minor));
-	WARN_ON(!disk->minors && !(disk->flags & GENHD_FL_EXT_DEVT));
+	WARN_ON(!disk->minors &&
+		!(disk->flags & (GENHD_FL_EXT_DEVT | GENHD_FL_HIDDEN)));
 
 	disk->flags |= GENHD_FL_UP;
 
@@ -648,22 +658,26 @@ void device_add_disk(struct device *parent, struct gendisk *disk)
 		WARN_ON(1);
 		return;
 	}
-	disk_to_dev(disk)->devt = devt;
-
-	/* ->major and ->first_minor aren't supposed to be
-	 * dereferenced from here on, but set them just in case.
-	 */
 	disk->major = MAJOR(devt);
 	disk->first_minor = MINOR(devt);
 
 	disk_alloc_events(disk);
 
-	/* Register BDI before referencing it from bdev */
-	bdi = disk->queue->backing_dev_info;
-	bdi_register_owner(bdi, disk_to_dev(disk));
-
-	blk_register_region(disk_devt(disk), disk->minors, NULL,
-			    exact_match, exact_lock, disk);
+	if (disk->flags & GENHD_FL_HIDDEN) {
+		/*
+		 * Don't let hidden disks show up in /proc/partitions,
+		 * and don't bother scanning for partitions either.
+		 */
+		disk->flags |= GENHD_FL_SUPPRESS_PARTITION_INFO;
+		disk->flags |= GENHD_FL_NO_PART_SCAN;
+	} else {
+		/* Register BDI before referencing it from bdev */
+		disk_to_dev(disk)->devt = devt;
+		bdi_register_owner(disk->queue->backing_dev_info,
+				disk_to_dev(disk));
+		blk_register_region(disk_devt(disk), disk->minors, NULL,
+				    exact_match, exact_lock, disk);
+	}
 	register_disk(parent, disk);
 	blk_register_queue(disk);
 
@@ -673,10 +687,6 @@ void device_add_disk(struct device *parent, struct gendisk *disk)
 	 */
 	WARN_ON_ONCE(!blk_get_queue(disk->queue));
 
-	retval = sysfs_create_link(&disk_to_dev(disk)->kobj, &bdi->dev->kobj,
-				   "bdi");
-	WARN_ON(retval);
-
 	disk_add_events(disk);
 	blk_integrity_add(disk);
 }
@@ -705,7 +715,8 @@ void del_gendisk(struct gendisk *disk)
 	set_capacity(disk, 0);
 	disk->flags &= ~GENHD_FL_UP;
 
-	sysfs_remove_link(&disk_to_dev(disk)->kobj, "bdi");
+	if (!(disk->flags & GENHD_FL_HIDDEN))
+		sysfs_remove_link(&disk_to_dev(disk)->kobj, "bdi");
 	if (disk->queue) {
 		/*
 		 * Unregister bdi before releasing device numbers (as they can
@@ -716,13 +727,15 @@ void del_gendisk(struct gendisk *disk)
 	} else {
 		WARN_ON(1);
 	}
-	blk_unregister_region(disk_devt(disk), disk->minors);
 
-	part_stat_set_all(&disk->part0, 0);
-	disk->part0.stamp = 0;
+	if (!(disk->flags & GENHD_FL_HIDDEN))
+		blk_unregister_region(disk_devt(disk), disk->minors);
 
 	kobject_put(disk->part0.holder_dir);
 	kobject_put(disk->slave_dir);
+
+	part_stat_set_all(&disk->part0, 0);
+	disk->part0.stamp = 0;
 	if (!sysfs_deprecated)
 		sysfs_remove_link(block_depr, dev_name(disk_to_dev(disk)));
 	pm_runtime_set_memalloc_noio(disk_to_dev(disk), false);
@@ -785,6 +798,10 @@ struct gendisk *get_gendisk(dev_t devt, int *partno)
 		spin_unlock_bh(&ext_devt_lock);
 	}
 
+	if (disk && unlikely(disk->flags & GENHD_FL_HIDDEN)) {
+		put_disk(disk);
+		disk = NULL;
+	}
 	return disk;
 }
 EXPORT_SYMBOL(get_gendisk);
@@ -1028,6 +1045,15 @@ static ssize_t disk_removable_show(struct device *dev,
 		       (disk->flags & GENHD_FL_REMOVABLE ? 1 : 0));
 }
 
+static ssize_t disk_hidden_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct gendisk *disk = dev_to_disk(dev);
+
+	return sprintf(buf, "%d\n",
+		       (disk->flags & GENHD_FL_HIDDEN ? 1 : 0));
+}
+
 static ssize_t disk_ro_show(struct device *dev,
 				   struct device_attribute *attr, char *buf)
 {
@@ -1065,6 +1091,7 @@ static ssize_t disk_discard_alignment_show(struct device *dev,
 static DEVICE_ATTR(range, S_IRUGO, disk_range_show, NULL);
 static DEVICE_ATTR(ext_range, S_IRUGO, disk_ext_range_show, NULL);
 static DEVICE_ATTR(removable, S_IRUGO, disk_removable_show, NULL);
+static DEVICE_ATTR(hidden, S_IRUGO, disk_hidden_show, NULL);
 static DEVICE_ATTR(ro, S_IRUGO, disk_ro_show, NULL);
 static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL);
 static DEVICE_ATTR(alignment_offset, S_IRUGO, disk_alignment_offset_show, NULL);
@@ -1089,6 +1116,7 @@ static struct attribute *disk_attrs[] = {
 	&dev_attr_range.attr,
 	&dev_attr_ext_range.attr,
 	&dev_attr_removable.attr,
+	&dev_attr_hidden.attr,
 	&dev_attr_ro.attr,
 	&dev_attr_size.attr,
 	&dev_attr_alignment_offset.attr,
diff --git a/block/ioctl.c b/block/ioctl.c
index 0de02ee67eed..1668506d8ed8 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -202,10 +202,16 @@ static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
 {
 	uint64_t range[2];
 	uint64_t start, len;
+	struct request_queue *q = bdev_get_queue(bdev);
+	struct address_space *mapping = bdev->bd_inode->i_mapping;
+
 
 	if (!(mode & FMODE_WRITE))
 		return -EBADF;
 
+	if (!blk_queue_discard(q))
+		return -EOPNOTSUPP;
+
 	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
 		return -EFAULT;
 
@@ -216,12 +222,12 @@ static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
 		return -EINVAL;
 	if (len & 511)
 		return -EINVAL;
-	start >>= 9;
-	len >>= 9;
 
-	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
+	if (start + len > i_size_read(bdev->bd_inode))
 		return -EINVAL;
-	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
+	truncate_inode_pages_range(mapping, start, start + len);
+	return blkdev_issue_discard(bdev, start >> 9, len >> 9,
+				    GFP_KERNEL, flags);
 }
 
 static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
@@ -437,11 +443,12 @@ static int blkdev_roset(struct block_device *bdev, fmode_t mode,
 {
 	int ret, n;
 
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
 	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
 	if (!is_unrecognized_ioctl(ret))
 		return ret;
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
 	if (get_user(n, (int __user *)arg))
 		return -EFAULT;
 	set_device_ro(bdev, n);
diff --git a/block/kyber-iosched.c b/block/kyber-iosched.c
index f58cab82105b..b4df317c2916 100644
--- a/block/kyber-iosched.c
+++ b/block/kyber-iosched.c
@@ -541,9 +541,17 @@ static int kyber_get_domain_token(struct kyber_queue_data *kqd,
 
 		/*
 		 * Try again in case a token was freed before we got on the wait
-		 * queue.
+		 * queue. The waker may have already removed the entry from the
+		 * wait queue, but list_del_init() is okay with that.
 		 */
 		nr = __sbitmap_queue_get(domain_tokens);
+		if (nr >= 0) {
+			unsigned long flags;
+
+			spin_lock_irqsave(&ws->wait.lock, flags);
+			list_del_init(&wait->entry);
+			spin_unlock_irqrestore(&ws->wait.lock, flags);
+		}
 	}
 	return nr;
 }
@@ -641,7 +649,7 @@ static bool kyber_has_work(struct blk_mq_hw_ctx *hctx)
 		if (!list_empty_careful(&khd->rqs[i]))
 			return true;
 	}
-	return false;
+	return sbitmap_any_bit_set(&hctx->ctx_map);
 }
 
 #define KYBER_LAT_SHOW_STORE(op)					\
diff --git a/block/mq-deadline.c b/block/mq-deadline.c
index a1cad4331edd..0179e484ec98 100644
--- a/block/mq-deadline.c
+++ b/block/mq-deadline.c
@@ -657,6 +657,7 @@ static struct elevator_type mq_deadline = {
 #endif
 	.elevator_attrs = deadline_attrs,
 	.elevator_name = "mq-deadline",
+	.elevator_alias = "deadline",
 	.elevator_owner = THIS_MODULE,
 };
 MODULE_ALIAS("mq-deadline-iosched");
diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c
index 7440de44dd85..edcfff974527 100644
--- a/block/scsi_ioctl.c
+++ b/block/scsi_ioctl.c
@@ -207,7 +207,7 @@ static void blk_set_cmd_filter_defaults(struct blk_cmd_filter *filter)
 	__set_bit(GPCMD_SET_READ_AHEAD, filter->write_ok);
 }
 
-int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm)
+int blk_verify_command(unsigned char *cmd, fmode_t mode)
 {
 	struct blk_cmd_filter *filter = &blk_default_cmd_filter;
 
@@ -220,7 +220,7 @@ int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm)
 		return 0;
 
 	/* Write-safe commands require a writable open */
-	if (test_bit(cmd[0], filter->write_ok) && has_write_perm)
+	if (test_bit(cmd[0], filter->write_ok) && (mode & FMODE_WRITE))
 		return 0;
 
 	return -EPERM;
@@ -234,7 +234,7 @@ static int blk_fill_sghdr_rq(struct request_queue *q, struct request *rq,
 
 	if (copy_from_user(req->cmd, hdr->cmdp, hdr->cmd_len))
 		return -EFAULT;
-	if (blk_verify_command(req->cmd, mode & FMODE_WRITE))
+	if (blk_verify_command(req->cmd, mode))
 		return -EPERM;
 
 	/*
@@ -469,7 +469,7 @@ int sg_scsi_ioctl(struct request_queue *q, struct gendisk *disk, fmode_t mode,
 	if (in_len && copy_from_user(buffer, sic->data + cmdlen, in_len))
 		goto error;
 
-	err = blk_verify_command(req->cmd, mode & FMODE_WRITE);
+	err = blk_verify_command(req->cmd, mode);
 	if (err)
 		goto error;
author	Linus Torvalds <torvalds@linux-foundation.org>	2017-11-14 15:32:19 -0800
committer	Linus Torvalds <torvalds@linux-foundation.org>	2017-11-14 15:32:19 -0800
commit	e2c5923c349c1738fe8fda980874d93f6fb2e5b6 (patch)
tree	b97a90170c45211bcc437761653aa8016c34afcd /block
parent	abc36be236358162202e86ad88616ff95a755101 (diff)
parent	a04b5de5050ab8b891128eb2c47a0916fe8622e1 (diff)
download	linux-e2c5923c349c1738fe8fda980874d93f6fb2e5b6.tar.bz2