Merge branch 'master' into mm-stable

14 files changed, 91 insertions, 111 deletions

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 0d46cb728bbf..e6d7e6b01a05 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -7046,6 +7046,7 @@ static void bfq_exit_queue(struct elevator_queue *e)
 	spin_unlock_irq(&bfqd->lock);
 #endif
 
+	blk_stat_disable_accounting(bfqd->queue);
 	wbt_enable_default(bfqd->queue);
 
 	kfree(bfqd);
@@ -7188,7 +7189,12 @@ 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);
 
+	/* We dispatch from request queue wide instead of hw queue */
+	blk_queue_flag_set(QUEUE_FLAG_SQ_SCHED, q);
+
 	wbt_disable_default(q);
+	blk_stat_enable_accounting(q);
+
 	return 0;
 
 out_free:
diff --git a/block/blk-core.c b/block/blk-core.c
index 06ff5bbfe8f6..27fb1357ad4b 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -322,19 +322,6 @@ void blk_cleanup_queue(struct request_queue *q)
 		blk_mq_exit_queue(q);
 	}
 
-	/*
-	 * In theory, request pool of sched_tags belongs to request queue.
-	 * However, the current implementation requires tag_set for freeing
-	 * requests, so free the pool now.
-	 *
-	 * Queue has become frozen, there can't be any in-queue requests, so
-	 * it is safe to free requests now.
-	 */
-	mutex_lock(&q->sysfs_lock);
-	if (q->elevator)
-		blk_mq_sched_free_rqs(q);
-	mutex_unlock(&q->sysfs_lock);
-
 	/* @q is and will stay empty, shutdown and put */
 	blk_put_queue(q);
 }
diff --git a/block/blk-ia-ranges.c b/block/blk-ia-ranges.c
index 56ed48d2954e..47c89e65b57f 100644
--- a/block/blk-ia-ranges.c
+++ b/block/blk-ia-ranges.c
@@ -144,7 +144,6 @@ int disk_register_independent_access_ranges(struct gendisk *disk,
 	}
 
 	for (i = 0; i < iars->nr_ia_ranges; i++) {
-		iars->ia_range[i].queue = q;
 		ret = kobject_init_and_add(&iars->ia_range[i].kobj,
 					   &blk_ia_range_ktype, &iars->kobj,
 					   "%d", i);
diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c
index 7e4136a60e1c..4d1ce9ef4318 100644
--- a/block/blk-mq-debugfs.c
+++ b/block/blk-mq-debugfs.c
@@ -711,11 +711,6 @@ void blk_mq_debugfs_register(struct request_queue *q)
 	}
 }
 
-void blk_mq_debugfs_unregister(struct request_queue *q)
-{
-	q->sched_debugfs_dir = NULL;
-}
-
 static void blk_mq_debugfs_register_ctx(struct blk_mq_hw_ctx *hctx,
 					struct blk_mq_ctx *ctx)
 {
@@ -746,6 +741,8 @@ void blk_mq_debugfs_register_hctx(struct request_queue *q,
 
 void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx)
 {
+	if (!hctx->queue->debugfs_dir)
+		return;
 	debugfs_remove_recursive(hctx->debugfs_dir);
 	hctx->sched_debugfs_dir = NULL;
 	hctx->debugfs_dir = NULL;
@@ -773,6 +770,8 @@ void blk_mq_debugfs_register_sched(struct request_queue *q)
 {
 	struct elevator_type *e = q->elevator->type;
 
+	lockdep_assert_held(&q->debugfs_mutex);
+
 	/*
 	 * If the parent directory has not been created yet, return, we will be
 	 * called again later on and the directory/files will be created then.
@@ -790,6 +789,8 @@ void blk_mq_debugfs_register_sched(struct request_queue *q)
 
 void blk_mq_debugfs_unregister_sched(struct request_queue *q)
 {
+	lockdep_assert_held(&q->debugfs_mutex);
+
 	debugfs_remove_recursive(q->sched_debugfs_dir);
 	q->sched_debugfs_dir = NULL;
 }
@@ -811,6 +812,10 @@ static const char *rq_qos_id_to_name(enum rq_qos_id id)
 
 void blk_mq_debugfs_unregister_rqos(struct rq_qos *rqos)
 {
+	lockdep_assert_held(&rqos->q->debugfs_mutex);
+
+	if (!rqos->q->debugfs_dir)
+		return;
 	debugfs_remove_recursive(rqos->debugfs_dir);
 	rqos->debugfs_dir = NULL;
 }
@@ -820,6 +825,8 @@ void blk_mq_debugfs_register_rqos(struct rq_qos *rqos)
 	struct request_queue *q = rqos->q;
 	const char *dir_name = rq_qos_id_to_name(rqos->id);
 
+	lockdep_assert_held(&q->debugfs_mutex);
+
 	if (rqos->debugfs_dir || !rqos->ops->debugfs_attrs)
 		return;
 
@@ -833,17 +840,13 @@ void blk_mq_debugfs_register_rqos(struct rq_qos *rqos)
 	debugfs_create_files(rqos->debugfs_dir, rqos, rqos->ops->debugfs_attrs);
 }
 
-void blk_mq_debugfs_unregister_queue_rqos(struct request_queue *q)
-{
-	debugfs_remove_recursive(q->rqos_debugfs_dir);
-	q->rqos_debugfs_dir = NULL;
-}
-
 void blk_mq_debugfs_register_sched_hctx(struct request_queue *q,
 					struct blk_mq_hw_ctx *hctx)
 {
 	struct elevator_type *e = q->elevator->type;
 
+	lockdep_assert_held(&q->debugfs_mutex);
+
 	/*
 	 * If the parent debugfs directory has not been created yet, return;
 	 * We will be called again later on with appropriate parent debugfs
@@ -863,6 +866,10 @@ void blk_mq_debugfs_register_sched_hctx(struct request_queue *q,
 
 void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx)
 {
+	lockdep_assert_held(&hctx->queue->debugfs_mutex);
+
+	if (!hctx->queue->debugfs_dir)
+		return;
 	debugfs_remove_recursive(hctx->sched_debugfs_dir);
 	hctx->sched_debugfs_dir = NULL;
 }
diff --git a/block/blk-mq-debugfs.h b/block/blk-mq-debugfs.h
index 69918f4170d6..9c7d4b6117d4 100644
--- a/block/blk-mq-debugfs.h
+++ b/block/blk-mq-debugfs.h
@@ -21,7 +21,6 @@ int __blk_mq_debugfs_rq_show(struct seq_file *m, struct request *rq);
 int blk_mq_debugfs_rq_show(struct seq_file *m, void *v);
 
 void blk_mq_debugfs_register(struct request_queue *q);
-void blk_mq_debugfs_unregister(struct request_queue *q);
 void blk_mq_debugfs_register_hctx(struct request_queue *q,
 				  struct blk_mq_hw_ctx *hctx);
 void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx);
@@ -36,16 +35,11 @@ void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx);
 
 void blk_mq_debugfs_register_rqos(struct rq_qos *rqos);
 void blk_mq_debugfs_unregister_rqos(struct rq_qos *rqos);
-void blk_mq_debugfs_unregister_queue_rqos(struct request_queue *q);
 #else
 static inline void blk_mq_debugfs_register(struct request_queue *q)
 {
 }
 
-static inline void blk_mq_debugfs_unregister(struct request_queue *q)
-{
-}
-
 static inline void blk_mq_debugfs_register_hctx(struct request_queue *q,
 						struct blk_mq_hw_ctx *hctx)
 {
@@ -87,10 +81,6 @@ static inline void blk_mq_debugfs_register_rqos(struct rq_qos *rqos)
 static inline void blk_mq_debugfs_unregister_rqos(struct rq_qos *rqos)
 {
 }
-
-static inline void blk_mq_debugfs_unregister_queue_rqos(struct request_queue *q)
-{
-}
 #endif
 
 #ifdef CONFIG_BLK_DEBUG_FS_ZONED
diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index 9e56a69422b6..a4f7c101b53b 100644
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -564,6 +564,7 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
 	int ret;
 
 	if (!e) {
+		blk_queue_flag_clear(QUEUE_FLAG_SQ_SCHED, q);
 		q->elevator = NULL;
 		q->nr_requests = q->tag_set->queue_depth;
 		return 0;
@@ -593,7 +594,9 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
 	if (ret)
 		goto err_free_map_and_rqs;
 
+	mutex_lock(&q->debugfs_mutex);
 	blk_mq_debugfs_register_sched(q);
+	mutex_unlock(&q->debugfs_mutex);
 
 	queue_for_each_hw_ctx(q, hctx, i) {
 		if (e->ops.init_hctx) {
@@ -606,7 +609,9 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
 				return ret;
 			}
 		}
+		mutex_lock(&q->debugfs_mutex);
 		blk_mq_debugfs_register_sched_hctx(q, hctx);
+		mutex_unlock(&q->debugfs_mutex);
 	}
 
 	return 0;
@@ -647,14 +652,21 @@ void blk_mq_exit_sched(struct request_queue *q, struct elevator_queue *e)
 	unsigned int flags = 0;
 
 	queue_for_each_hw_ctx(q, hctx, i) {
+		mutex_lock(&q->debugfs_mutex);
 		blk_mq_debugfs_unregister_sched_hctx(hctx);
+		mutex_unlock(&q->debugfs_mutex);
+
 		if (e->type->ops.exit_hctx && hctx->sched_data) {
 			e->type->ops.exit_hctx(hctx, i);
 			hctx->sched_data = NULL;
 		}
 		flags = hctx->flags;
 	}
+
+	mutex_lock(&q->debugfs_mutex);
 	blk_mq_debugfs_unregister_sched(q);
+	mutex_unlock(&q->debugfs_mutex);
+
 	if (e->type->ops.exit_sched)
 		e->type->ops.exit_sched(e);
 	blk_mq_sched_tags_teardown(q, flags);
diff --git a/block/blk-mq.c b/block/blk-mq.c
index e9bf950983c7..93d9d60980fb 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -579,6 +579,8 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
 	if (!blk_mq_hw_queue_mapped(data.hctx))
 		goto out_queue_exit;
 	cpu = cpumask_first_and(data.hctx->cpumask, cpu_online_mask);
+	if (cpu >= nr_cpu_ids)
+		goto out_queue_exit;
 	data.ctx = __blk_mq_get_ctx(q, cpu);
 
 	if (!q->elevator)
@@ -2141,20 +2143,6 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
 EXPORT_SYMBOL(blk_mq_run_hw_queue);
 
 /*
- * Is the request queue handled by an IO scheduler that does not respect
- * hardware queues when dispatching?
- */
-static bool blk_mq_has_sqsched(struct request_queue *q)
-{
-	struct elevator_queue *e = q->elevator;
-
-	if (e && e->type->ops.dispatch_request &&
-	    !(e->type->elevator_features & ELEVATOR_F_MQ_AWARE))
-		return true;
-	return false;
-}
-
-/*
  * Return prefered queue to dispatch from (if any) for non-mq aware IO
  * scheduler.
  */
@@ -2186,7 +2174,7 @@ void blk_mq_run_hw_queues(struct request_queue *q, bool async)
 	unsigned long i;
 
 	sq_hctx = NULL;
-	if (blk_mq_has_sqsched(q))
+	if (blk_queue_sq_sched(q))
 		sq_hctx = blk_mq_get_sq_hctx(q);
 	queue_for_each_hw_ctx(q, hctx, i) {
 		if (blk_mq_hctx_stopped(hctx))
@@ -2214,7 +2202,7 @@ void blk_mq_delay_run_hw_queues(struct request_queue *q, unsigned long msecs)
 	unsigned long i;
 
 	sq_hctx = NULL;
-	if (blk_mq_has_sqsched(q))
+	if (blk_queue_sq_sched(q))
 		sq_hctx = blk_mq_get_sq_hctx(q);
 	queue_for_each_hw_ctx(q, hctx, i) {
 		if (blk_mq_hctx_stopped(hctx))
@@ -2777,15 +2765,20 @@ static inline struct request *blk_mq_get_cached_request(struct request_queue *q,
 		return NULL;
 	}
 
-	rq_qos_throttle(q, *bio);
-
 	if (blk_mq_get_hctx_type((*bio)->bi_opf) != rq->mq_hctx->type)
 		return NULL;
 	if (op_is_flush(rq->cmd_flags) != op_is_flush((*bio)->bi_opf))
 		return NULL;
 
-	rq->cmd_flags = (*bio)->bi_opf;
+	/*
+	 * If any qos ->throttle() end up blocking, we will have flushed the
+	 * plug and hence killed the cached_rq list as well. Pop this entry
+	 * before we throttle.
+	 */
 	plug->cached_rq = rq_list_next(rq);
+	rq_qos_throttle(q, *bio);
+
+	rq->cmd_flags = (*bio)->bi_opf;
 	INIT_LIST_HEAD(&rq->queuelist);
 	return rq;
 }
@@ -3443,8 +3436,9 @@ static void blk_mq_exit_hctx(struct request_queue *q,
 	if (blk_mq_hw_queue_mapped(hctx))
 		blk_mq_tag_idle(hctx);
 
-	blk_mq_clear_flush_rq_mapping(set->tags[hctx_idx],
-			set->queue_depth, flush_rq);
+	if (blk_queue_init_done(q))
+		blk_mq_clear_flush_rq_mapping(set->tags[hctx_idx],
+				set->queue_depth, flush_rq);
 	if (set->ops->exit_request)
 		set->ops->exit_request(set, flush_rq, hctx_idx);
 
@@ -4438,12 +4432,14 @@ static bool blk_mq_elv_switch_none(struct list_head *head,
 	if (!qe)
 		return false;
 
+	/* q->elevator needs protection from ->sysfs_lock */
+	mutex_lock(&q->sysfs_lock);
+
 	INIT_LIST_HEAD(&qe->node);
 	qe->q = q;
 	qe->type = q->elevator->type;
 	list_add(&qe->node, head);
 
-	mutex_lock(&q->sysfs_lock);
 	/*
 	 * After elevator_switch_mq, the previous elevator_queue will be
 	 * released by elevator_release. The reference of the io scheduler
diff --git a/block/blk-rq-qos.c b/block/blk-rq-qos.c
index e83af7bc7591..d3a75693adbf 100644
--- a/block/blk-rq-qos.c
+++ b/block/blk-rq-qos.c
@@ -294,8 +294,6 @@ void rq_qos_wait(struct rq_wait *rqw, void *private_data,
 
 void rq_qos_exit(struct request_queue *q)
 {
-	blk_mq_debugfs_unregister_queue_rqos(q);
-
 	while (q->rq_qos) {
 		struct rq_qos *rqos = q->rq_qos;
 		q->rq_qos = rqos->next;
diff --git a/block/blk-rq-qos.h b/block/blk-rq-qos.h
index 68267007da1c..0e46052b018a 100644
--- a/block/blk-rq-qos.h
+++ b/block/blk-rq-qos.h
@@ -104,8 +104,11 @@ static inline void rq_qos_add(struct request_queue *q, struct rq_qos *rqos)
 
 	blk_mq_unfreeze_queue(q);
 
-	if (rqos->ops->debugfs_attrs)
+	if (rqos->ops->debugfs_attrs) {
+		mutex_lock(&q->debugfs_mutex);
 		blk_mq_debugfs_register_rqos(rqos);
+		mutex_unlock(&q->debugfs_mutex);
+	}
 }
 
 static inline void rq_qos_del(struct request_queue *q, struct rq_qos *rqos)
@@ -129,7 +132,9 @@ static inline void rq_qos_del(struct request_queue *q, struct rq_qos *rqos)
 
 	blk_mq_unfreeze_queue(q);
 
+	mutex_lock(&q->debugfs_mutex);
 	blk_mq_debugfs_unregister_rqos(rqos);
+	mutex_unlock(&q->debugfs_mutex);
 }
 
 typedef bool (acquire_inflight_cb_t)(struct rq_wait *rqw, void *private_data);
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 88bd41d4cb59..9b905e9443e4 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -779,14 +779,6 @@ static void blk_release_queue(struct kobject *kobj)
 	if (queue_is_mq(q))
 		blk_mq_release(q);
 
-	blk_trace_shutdown(q);
-	mutex_lock(&q->debugfs_mutex);
-	debugfs_remove_recursive(q->debugfs_dir);
-	mutex_unlock(&q->debugfs_mutex);
-
-	if (queue_is_mq(q))
-		blk_mq_debugfs_unregister(q);
-
 	bioset_exit(&q->bio_split);
 
 	if (blk_queue_has_srcu(q))
@@ -836,17 +828,16 @@ int blk_register_queue(struct gendisk *disk)
 		goto unlock;
 	}
 
+	if (queue_is_mq(q))
+		__blk_mq_register_dev(dev, q);
+	mutex_lock(&q->sysfs_lock);
+
 	mutex_lock(&q->debugfs_mutex);
 	q->debugfs_dir = debugfs_create_dir(kobject_name(q->kobj.parent),
 					    blk_debugfs_root);
-	mutex_unlock(&q->debugfs_mutex);
-
-	if (queue_is_mq(q)) {
-		__blk_mq_register_dev(dev, q);
+	if (queue_is_mq(q))
 		blk_mq_debugfs_register(q);
-	}
-
-	mutex_lock(&q->sysfs_lock);
+	mutex_unlock(&q->debugfs_mutex);
 
 	ret = disk_register_independent_access_ranges(disk, NULL);
 	if (ret)
@@ -948,8 +939,15 @@ void blk_unregister_queue(struct gendisk *disk)
 	/* Now that we've deleted all child objects, we can delete the queue. */
 	kobject_uevent(&q->kobj, KOBJ_REMOVE);
 	kobject_del(&q->kobj);
-
 	mutex_unlock(&q->sysfs_dir_lock);
 
+	mutex_lock(&q->debugfs_mutex);
+	blk_trace_shutdown(q);
+	debugfs_remove_recursive(q->debugfs_dir);
+	q->debugfs_dir = NULL;
+	q->sched_debugfs_dir = NULL;
+	q->rqos_debugfs_dir = NULL;
+	mutex_unlock(&q->debugfs_mutex);
+
 	kobject_put(&disk_to_dev(disk)->kobj);
 }
diff --git a/block/genhd.c b/block/genhd.c
index 27205ae47d59..278227ba1d53 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -623,6 +623,7 @@ void del_gendisk(struct gendisk *disk)
 	 * Prevent new I/O from crossing bio_queue_enter().
 	 */
 	blk_queue_start_drain(q);
+	blk_mq_freeze_queue_wait(q);
 
 	if (!(disk->flags & GENHD_FL_HIDDEN)) {
 		sysfs_remove_link(&disk_to_dev(disk)->kobj, "bdi");
@@ -646,12 +647,21 @@ void del_gendisk(struct gendisk *disk)
 	pm_runtime_set_memalloc_noio(disk_to_dev(disk), false);
 	device_del(disk_to_dev(disk));
 
-	blk_mq_freeze_queue_wait(q);
-
 	blk_throtl_cancel_bios(disk->queue);
 
 	blk_sync_queue(q);
 	blk_flush_integrity();
+	blk_mq_cancel_work_sync(q);
+
+	blk_mq_quiesce_queue(q);
+	if (q->elevator) {
+		mutex_lock(&q->sysfs_lock);
+		elevator_exit(q);
+		mutex_unlock(&q->sysfs_lock);
+	}
+	rq_qos_exit(q);
+	blk_mq_unquiesce_queue(q);
+
 	/*
 	 * Allow using passthrough request again after the queue is torn down.
 	 */
@@ -1120,31 +1130,6 @@ static const struct attribute_group *disk_attr_groups[] = {
 	NULL
 };
 
-static void disk_release_mq(struct request_queue *q)
-{
-	blk_mq_cancel_work_sync(q);
-
-	/*
-	 * There can't be any non non-passthrough bios in flight here, but
-	 * requests stay around longer, including passthrough ones so we
-	 * still need to freeze the queue here.
-	 */
-	blk_mq_freeze_queue(q);
-
-	/*
-	 * Since the I/O scheduler exit code may access cgroup information,
-	 * perform I/O scheduler exit before disassociating from the block
-	 * cgroup controller.
-	 */
-	if (q->elevator) {
-		mutex_lock(&q->sysfs_lock);
-		elevator_exit(q);
-		mutex_unlock(&q->sysfs_lock);
-	}
-	rq_qos_exit(q);
-	__blk_mq_unfreeze_queue(q, true);
-}
-
 /**
  * disk_release - releases all allocated resources of the gendisk
  * @dev: the device representing this disk
@@ -1166,9 +1151,6 @@ static void disk_release(struct device *dev)
 	might_sleep();
 	WARN_ON_ONCE(disk_live(disk));
 
-	if (queue_is_mq(disk->queue))
-		disk_release_mq(disk->queue);
-
 	blkcg_exit_queue(disk->queue);
 
 	disk_release_events(disk);
diff --git a/block/holder.c b/block/holder.c
index 8d750281a1cd..5283bc804cc1 100644
--- a/block/holder.c
+++ b/block/holder.c
@@ -79,10 +79,6 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk)
 
 	WARN_ON_ONCE(!bdev->bd_holder);
 
-	/* FIXME: remove the following once add_disk() handles errors */
-	if (WARN_ON(!bdev->bd_holder_dir))
-		goto out_unlock;
-
 	holder = bd_find_holder_disk(bdev, disk);
 	if (holder) {
 		holder->refcnt++;
diff --git a/block/kyber-iosched.c b/block/kyber-iosched.c
index 70ff2a599ef6..8f7c745b4a57 100644
--- a/block/kyber-iosched.c
+++ b/block/kyber-iosched.c
@@ -421,6 +421,8 @@ static int kyber_init_sched(struct request_queue *q, struct elevator_type *e)
 
 	blk_stat_enable_accounting(q);
 
+	blk_queue_flag_clear(QUEUE_FLAG_SQ_SCHED, q);
+
 	eq->elevator_data = kqd;
 	q->elevator = eq;
 
@@ -1033,7 +1035,6 @@ static struct elevator_type kyber_sched = {
 #endif
 	.elevator_attrs = kyber_sched_attrs,
 	.elevator_name = "kyber",
-	.elevator_features = ELEVATOR_F_MQ_AWARE,
 	.elevator_owner = THIS_MODULE,
 };
 
diff --git a/block/mq-deadline.c b/block/mq-deadline.c
index 6ed602b2f80a..1a9e835e816c 100644
--- a/block/mq-deadline.c
+++ b/block/mq-deadline.c
@@ -642,6 +642,9 @@ static int dd_init_sched(struct request_queue *q, struct elevator_type *e)
 	spin_lock_init(&dd->lock);
 	spin_lock_init(&dd->zone_lock);
 
+	/* We dispatch from request queue wide instead of hw queue */
+	blk_queue_flag_set(QUEUE_FLAG_SQ_SCHED, q);
+
 	q->elevator = eq;
 	return 0;