blk-mq.c

/*
 * Block multiqueue core code
 *
 * Copyright (C) 2013-2014 Jens Axboe
 * Copyright (C) 2013-2014 Christoph Hellwig
 */
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/backing-dev.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/mm.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/workqueue.h>
#include <linux/smp.h>
#include <linux/llist.h>
#include <linux/list_sort.h>
#include <linux/cpu.h>
#include <linux/cache.h>
#include <linux/sched/sysctl.h>
#include <linux/delay.h>
#include <linux/crash_dump.h>

#include <trace/events/block.h>

#include <linux/blk-mq.h>
#include "blk.h"
#include "blk-mq.h"
#include "blk-mq-tag.h"

static DEFINE_MUTEX(all_q_mutex);
static LIST_HEAD(all_q_list);

static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx);

/*
 * Check if any of the ctx's have pending work in this hardware queue
 */
static bool blk_mq_hctx_has_pending(struct blk_mq_hw_ctx *hctx)
{
	unsigned int i;

	for (i = 0; i < hctx->ctx_map.map_size; i++)
		if (hctx->ctx_map.map[i].word)
			return true;

	return false;
}

static inline struct blk_align_bitmap *get_bm(struct blk_mq_hw_ctx *hctx,
					      struct blk_mq_ctx *ctx)
{
	return &hctx->ctx_map.map[ctx->index_hw / hctx->ctx_map.bits_per_word];
}

#define CTX_TO_BIT(hctx, ctx)	\
	((ctx)->index_hw & ((hctx)->ctx_map.bits_per_word - 1))

/*
 * Mark this ctx as having pending work in this hardware queue
 */
static void blk_mq_hctx_mark_pending(struct blk_mq_hw_ctx *hctx,
				     struct blk_mq_ctx *ctx)
{
	struct blk_align_bitmap *bm = get_bm(hctx, ctx);

	if (!test_bit(CTX_TO_BIT(hctx, ctx), &bm->word))
		set_bit(CTX_TO_BIT(hctx, ctx), &bm->word);
}

static void blk_mq_hctx_clear_pending(struct blk_mq_hw_ctx *hctx,
				      struct blk_mq_ctx *ctx)
{
	struct blk_align_bitmap *bm = get_bm(hctx, ctx);

	clear_bit(CTX_TO_BIT(hctx, ctx), &bm->word);
}

static int blk_mq_queue_enter(struct request_queue *q)
{
	while (true) {
		int ret;

		if (percpu_ref_tryget_live(&q->mq_usage_counter))
			return 0;

		ret = wait_event_interruptible(q->mq_freeze_wq,
				!q->mq_freeze_depth || blk_queue_dying(q));
		if (blk_queue_dying(q))
			return -ENODEV;
		if (ret)
			return ret;
	}
}

static void blk_mq_queue_exit(struct request_queue *q)
{
	percpu_ref_put(&q->mq_usage_counter);
}

static void blk_mq_usage_counter_release(struct percpu_ref *ref)
{
	struct request_queue *q =
		container_of(ref, struct request_queue, mq_usage_counter);

	wake_up_all(&q->mq_freeze_wq);
}

void blk_mq_freeze_queue_start(struct request_queue *q)
{
	bool freeze;

	spin_lock_irq(q->queue_lock);
	freeze = !q->mq_freeze_depth++;
	spin_unlock_irq(q->queue_lock);

	if (freeze) {
		percpu_ref_kill(&q->mq_usage_counter);
		blk_mq_run_queues(q, false);
	}
}
EXPORT_SYMBOL_GPL(blk_mq_freeze_queue_start);

static void blk_mq_freeze_queue_wait(struct request_queue *q)
{
	wait_event(q->mq_freeze_wq, percpu_ref_is_zero(&q->mq_usage_counter));
}

/*
 * Guarantee no request is in use, so we can change any data structure of
 * the queue afterward.
 */
void blk_mq_freeze_queue(struct request_queue *q)
{
	blk_mq_freeze_queue_start(q);
	blk_mq_freeze_queue_wait(q);
}

void blk_mq_unfreeze_queue(struct request_queue *q)
{
	bool wake;

	spin_lock_irq(q->queue_lock);
	wake = !--q->mq_freeze_depth;
	WARN_ON_ONCE(q->mq_freeze_depth < 0);
	spin_unlock_irq(q->queue_lock);
	if (wake) {
		percpu_ref_reinit(&q->mq_usage_counter);
		wake_up_all(&q->mq_freeze_wq);
	}
}
EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue);

void blk_mq_wake_waiters(struct request_queue *q)
{
	struct blk_mq_hw_ctx *hctx;
	unsigned int i;

	queue_for_each_hw_ctx(q, hctx, i)
		if (blk_mq_hw_queue_mapped(hctx))
			blk_mq_tag_wakeup_all(hctx->tags, true);
}

bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
{
	return blk_mq_has_free_tags(hctx->tags);
}
EXPORT_SYMBOL(blk_mq_can_queue);

static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx,
			       struct request *rq, unsigned int rw_flags)
{
	if (blk_queue_io_stat(q))
		rw_flags |= REQ_IO_STAT;

	INIT_LIST_HEAD(&rq->queuelist);
	/* csd/requeue_work/fifo_time is initialized before use */
	rq->q = q;
	rq->mq_ctx = ctx;
	rq->cmd_flags |= rw_flags;
	/* do not touch atomic flags, it needs atomic ops against the timer */
	rq->cpu = -1;
	INIT_HLIST_NODE(&rq->hash);
	RB_CLEAR_NODE(&rq->rb_node);
	rq->rq_disk = NULL;
	rq->part = NULL;
	rq->start_time = jiffies;
#ifdef CONFIG_BLK_CGROUP
	rq->rl = NULL;
	set_start_time_ns(rq);
	rq->io_start_time_ns = 0;
#endif
	rq->nr_phys_segments = 0;
#if defined(CONFIG_BLK_DEV_INTEGRITY)
	rq->nr_integrity_segments = 0;
#endif
	rq->special = NULL;
	/* tag was already set */
	rq->errors = 0;

	rq->cmd = rq->__cmd;

	rq->extra_len = 0;
	rq->sense_len = 0;
	rq->resid_len = 0;
	rq->sense = NULL;

	INIT_LIST_HEAD(&rq->timeout_list);
	rq->timeout = 0;

	rq->end_io = NULL;
	rq->end_io_data = NULL;
	rq->next_rq = NULL;

	ctx->rq_dispatched[rw_is_sync(rw_flags)]++;
}

static struct request *
__blk_mq_alloc_request(struct blk_mq_alloc_data *data, int rw)
{
	struct request *rq;
	unsigned int tag;

	tag = blk_mq_get_tag(data);
	if (tag != BLK_MQ_TAG_FAIL) {
		rq = data->hctx->tags->rqs[tag];

		if (blk_mq_tag_busy(data->hctx)) {
			rq->cmd_flags = REQ_MQ_INFLIGHT;
			atomic_inc(&data->hctx->nr_active);
		}

		rq->tag = tag;
		blk_mq_rq_ctx_init(data->q, data->ctx, rq, rw);
		return rq;
	}

	return NULL;
}

struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp,
		bool reserved)
{
	struct blk_mq_ctx *ctx;
	struct blk_mq_hw_ctx *hctx;
	struct request *rq;
	struct blk_mq_alloc_data alloc_data;
	int ret;

	ret = blk_mq_queue_enter(q);
	if (ret)
		return ERR_PTR(ret);

	ctx = blk_mq_get_ctx(q);
	hctx = q->mq_ops->map_queue(q, ctx->cpu);
	blk_mq_set_alloc_data(&alloc_data, q, gfp & ~__GFP_WAIT,
			reserved, ctx, hctx);

	rq = __blk_mq_alloc_request(&alloc_data, rw);
	if (!rq && (gfp & __GFP_WAIT)) {
		__blk_mq_run_hw_queue(hctx);
		blk_mq_put_ctx(ctx);

		ctx = blk_mq_get_ctx(q);
		hctx = q->mq_ops->map_queue(q, ctx->cpu);
		blk_mq_set_alloc_data(&alloc_data, q, gfp, reserved, ctx,
				hctx);
		rq =  __blk_mq_alloc_request(&alloc_data, rw);
		ctx = alloc_data.ctx;
	}
	blk_mq_put_ctx(ctx);
	if (!rq) {
		blk_mq_queue_exit(q);
		return ERR_PTR(-EWOULDBLOCK);
	}
	return rq;
}
EXPORT_SYMBOL(blk_mq_alloc_request);

static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx,
				  struct blk_mq_ctx *ctx, struct request *rq)
{
	const int tag = rq->tag;
	struct request_queue *q = rq->q;

	if (rq->cmd_flags & REQ_MQ_INFLIGHT)
		atomic_dec(&hctx->nr_active);
	rq->cmd_flags = 0;

	clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags);
	blk_mq_put_tag(hctx, tag, &ctx->last_tag);
	blk_mq_queue_exit(q);
}

void blk_mq_free_hctx_request(struct blk_mq_hw_ctx *hctx, struct request *rq)
{
	struct blk_mq_ctx *ctx = rq->mq_ctx;

	ctx->rq_completed[rq_is_sync(rq)]++;
	__blk_mq_free_request(hctx, ctx, rq);

}
EXPORT_SYMBOL_GPL(blk_mq_free_hctx_request);

void blk_mq_free_request(struct request *rq)
{
	struct blk_mq_hw_ctx *hctx;
	struct request_queue *q = rq->q;

	hctx = q->mq_ops->map_queue(q, rq->mq_ctx->cpu);
	blk_mq_free_hctx_request(hctx, rq);
}
EXPORT_SYMBOL_GPL(blk_mq_free_request);

inline void __blk_mq_end_request(struct request *rq, int error)
{
	blk_account_io_done(rq);

	if (rq->end_io) {
		rq->end_io(rq, error);
	} else {
		if (unlikely(blk_bidi_rq(rq)))
			blk_mq_free_request(rq->next_rq);
		blk_mq_free_request(rq);
	}
}
EXPORT_SYMBOL(__blk_mq_end_request);

void blk_mq_end_request(struct request *rq, int error)
{
	if (blk_update_request(rq, error, blk_rq_bytes(rq)))
		BUG();
	__blk_mq_end_request(rq, error);
}
EXPORT_SYMBOL(blk_mq_end_request);

static void __blk_mq_complete_request_remote(void *data)
{
	struct request *rq = data;

	rq->q->softirq_done_fn(rq);
}

static void blk_mq_ipi_complete_request(struct request *rq)
{
	struct blk_mq_ctx *ctx = rq->mq_ctx;
	bool shared = false;
	int cpu;

	if (!test_bit(QUEUE_FLAG_SAME_COMP, &rq->q->queue_flags)) {
		rq->q->softirq_done_fn(rq);
		return;
	}

	cpu = get_cpu();
	if (!test_bit(QUEUE_FLAG_SAME_FORCE, &rq->q->queue_flags))
		shared = cpus_share_cache(cpu, ctx->cpu);

	if (cpu != ctx->cpu && !shared && cpu_online(ctx->cpu)) {
		rq->csd.func = __blk_mq_complete_request_remote;
		rq->csd.info = rq;
		rq->csd.flags = 0;
		smp_call_function_single_async(ctx->cpu, &rq->csd);
	} else {
		rq->q->softirq_done_fn(rq);
	}
	put_cpu();
}

void __blk_mq_complete_request(struct request *rq)
{
	struct request_queue *q = rq->q;

	if (!q->softirq_done_fn)
		blk_mq_end_request(rq, rq->errors);
	else
		blk_mq_ipi_complete_request(rq);
}

/**
 * blk_mq_complete_request - end I/O on a request
 * @rq:		the request being processed
 *
 * Description:
 *	Ends all I/O on a request. It does not handle partial completions.
 *	The actual completion happens out-of-order, through a IPI handler.
 **/
void blk_mq_complete_request(struct request *rq)
{
	struct request_queue *q = rq->q;

	if (unlikely(blk_should_fake_timeout(q)))
		return;
	if (!blk_mark_rq_complete(rq))
		__blk_mq_complete_request(rq);
}
EXPORT_SYMBOL(blk_mq_complete_request);

void blk_mq_start_request(struct request *rq)
{
	struct request_queue *q = rq->q;

	trace_block_rq_issue(q, rq);

	rq->resid_len = blk_rq_bytes(rq);
	if (unlikely(blk_bidi_rq(rq)))
		rq->next_rq->resid_len = blk_rq_bytes(rq->next_rq);

	blk_add_timer(rq);

	/*
	 * Ensure that ->deadline is visible before set the started
	 * flag and clear the completed flag.
	 */
	smp_mb__before_atomic();

	/*
	 * 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.
	 */
	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))
		clear_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags);

	if (q->dma_drain_size && blk_rq_bytes(rq)) {
		/*
		 * Make sure space for the drain appears.  We know we can do
		 * this because max_hw_segments has been adjusted to be one
		 * fewer than the device can handle.
		 */
		rq->nr_phys_segments++;
	}
}
EXPORT_SYMBOL(blk_mq_start_request);

static void __blk_mq_requeue_request(struct request *rq)
{
	struct request_queue *q = rq->q;

	trace_block_rq_requeue(q, rq);

	if (test_and_clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags)) {
		if (q->dma_drain_size && blk_rq_bytes(rq))
			rq->nr_phys_segments--;
	}
}

void blk_mq_requeue_request(struct request *rq)
{
	__blk_mq_requeue_request(rq);

	BUG_ON(blk_queued_rq(rq));
	blk_mq_add_to_requeue_list(rq, true);
}
EXPORT_SYMBOL(blk_mq_requeue_request);

static void blk_mq_requeue_work(struct work_struct *work)
{
	struct request_queue *q =
		container_of(work, struct request_queue, requeue_work);
	LIST_HEAD(rq_list);
	struct request *rq, *next;
	unsigned long flags;

	spin_lock_irqsave(&q->requeue_lock, flags);
	list_splice_init(&q->requeue_list, &rq_list);
	spin_unlock_irqrestore(&q->requeue_lock, flags);

	list_for_each_entry_safe(rq, next, &rq_list, queuelist) {
		if (!(rq->cmd_flags & REQ_SOFTBARRIER))
			continue;

		rq->cmd_flags &= ~REQ_SOFTBARRIER;
		list_del_init(&rq->queuelist);
		blk_mq_insert_request(rq, true, false, false);
	}

	while (!list_empty(&rq_list)) {
		rq = list_entry(rq_list.next, struct request, queuelist);
		list_del_init(&rq->queuelist);
		blk_mq_insert_request(rq, false, false, false);
	}

	/*
	 * Use the start variant of queue running here, so that running
	 * the requeue work will kick stopped queues.
	 */
	blk_mq_start_hw_queues(q);
}

void blk_mq_add_to_requeue_list(struct request *rq, bool at_head)
{
	struct request_queue *q = rq->q;
	unsigned long flags;

	/*
	 * We abuse this flag that is otherwise used by the I/O scheduler to
	 * request head insertation from the workqueue.
	 */
	BUG_ON(rq->cmd_flags & REQ_SOFTBARRIER);

	spin_lock_irqsave(&q->requeue_lock, flags);
	if (at_head) {
		rq->cmd_flags |= REQ_SOFTBARRIER;
		list_add(&rq->queuelist, &q->requeue_list);
	} else {
		list_add_tail(&rq->queuelist, &q->requeue_list);
	}
	spin_unlock_irqrestore(&q->requeue_lock, flags);
}
EXPORT_SYMBOL(blk_mq_add_to_requeue_list);

void blk_mq_kick_requeue_list(struct request_queue *q)
{
	kblockd_schedule_work(&q->requeue_work);
}
EXPORT_SYMBOL(blk_mq_kick_requeue_list);

static inline bool is_flush_request(struct request *rq,
		struct blk_flush_queue *fq, unsigned int tag)
{
	return ((rq->cmd_flags & REQ_FLUSH_SEQ) &&
			fq->flush_rq->tag == tag);
}

struct request *blk_mq_tag_to_rq(struct blk_mq_tags *tags, unsigned int tag)
{
	struct request *rq = tags->rqs[tag];
	/* mq_ctx of flush rq is always cloned from the corresponding req */
	struct blk_flush_queue *fq = blk_get_flush_queue(rq->q, rq->mq_ctx);

	if (!is_flush_request(rq, fq, tag))
		return rq;

	return fq->flush_rq;
}
EXPORT_SYMBOL(blk_mq_tag_to_rq);

struct blk_mq_timeout_data {
	unsigned long next;
	unsigned int next_set;
};

void blk_mq_rq_timed_out(struct request *req, bool reserved)
{
	struct blk_mq_ops *ops = req->q->mq_ops;
	enum blk_eh_timer_return ret = BLK_EH_RESET_TIMER;

	/*
	 * We know that complete is set at this point. If STARTED isn't set
	 * anymore, then the request isn't active and the "timeout" should
	 * just be ignored. This can happen due to the bitflag ordering.
	 * Timeout first checks if STARTED is set, and if it is, assumes
	 * the request is active. But if we race with completion, then
	 * we both flags will get cleared. So check here again, and ignore
	 * a timeout event with a request that isn't active.
	 */
	if (!test_bit(REQ_ATOM_STARTED, &req->atomic_flags))
		return;

	if (ops->timeout)
		ret = ops->timeout(req, reserved);

	switch (ret) {
	case BLK_EH_HANDLED:
		__blk_mq_complete_request(req);
		break;
	case BLK_EH_RESET_TIMER:
		blk_add_timer(req);
		blk_clear_rq_complete(req);
		break;
	case BLK_EH_NOT_HANDLED:
		break;
	default:
		printk(KERN_ERR "block: bad eh return: %d\n", ret);
		break;
	}
}
		
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;

	if (!test_bit(REQ_ATOM_STARTED, &rq->atomic_flags))
		return;

	if (time_after_eq(jiffies, rq->deadline)) {
		if (!blk_mark_rq_complete(rq))
			blk_mq_rq_timed_out(rq, reserved);
	} else if (!data->next_set || time_after(data->next, rq->deadline)) {
		data->next = rq->deadline;
		data->next_set = 1;
	}
}

static void blk_mq_rq_timer(unsigned long priv)
{
	struct request_queue *q = (struct request_queue *)priv;
	struct blk_mq_timeout_data data = {
		.next		= 0,
		.next_set	= 0,
	};
	struct blk_mq_hw_ctx *hctx;
	int i;

	queue_for_each_hw_ctx(q, hctx, i) {
		/*
		 * If not software queues are currently mapped to this
		 * hardware queue, there's nothing to check
		 */
		if (!blk_mq_hw_queue_mapped(hctx))
			continue;

		blk_mq_tag_busy_iter(hctx, blk_mq_check_expired, &data);
	}

	if (data.next_set) {
		data.next = blk_rq_timeout(round_jiffies_up(data.next));
		mod_timer(&q->timeout, data.next);
	} else {
		queue_for_each_hw_ctx(q, hctx, i)
			blk_mq_tag_idle(hctx);
	}
}

/*
 * Reverse check our software queue for entries that we could potentially
 * merge with. Currently includes a hand-wavy stop count of 8, to not spend
 * too much time checking for merges.
 */
static bool blk_mq_attempt_merge(struct request_queue *q,
				 struct blk_mq_ctx *ctx, struct bio *bio)
{
	struct request *rq;
	int checked = 8;

	list_for_each_entry_reverse(rq, &ctx->rq_list, queuelist) {
		int el_ret;

		if (!checked--)
			break;

		if (!blk_rq_merge_ok(rq, bio))
			continue;

		el_ret = blk_try_merge(rq, bio);
		if (el_ret == ELEVATOR_BACK_MERGE) {
			if (bio_attempt_back_merge(q, rq, bio)) {
				ctx->rq_merged++;
				return true;
			}
			break;
		} else if (el_ret == ELEVATOR_FRONT_MERGE) {
			if (bio_attempt_front_merge(q, rq, bio)) {
				ctx->rq_merged++;
				return true;
			}
			break;
		}
	}

	return false;
}

/*
 * Process software queues that have been marked busy, splicing them
 * to the for-dispatch
 */
static void flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list)
{
	struct blk_mq_ctx *ctx;
	int i;

	for (i = 0; i < hctx->ctx_map.map_size; i++) {
		struct blk_align_bitmap *bm = &hctx->ctx_map.map[i];
		unsigned int off, bit;

		if (!bm->word)
			continue;

		bit = 0;
		off = i * hctx->ctx_map.bits_per_word;
		do {
			bit = find_next_bit(&bm->word, bm->depth, bit);
			if (bit >= bm->depth)
				break;

			ctx = hctx->ctxs[bit + off];
			clear_bit(bit, &bm->word);
			spin_lock(&ctx->lock);
			list_splice_tail_init(&ctx->rq_list, list);
			spin_unlock(&ctx->lock);

			bit++;
		} while (1);
	}
}

/*
 * Run this hardware queue, pulling any software queues mapped to it in.
 * Note that this function currently has various problems around ordering
 * of IO. In particular, we'd like FIFO behaviour on handling existing
 * items on the hctx->dispatch list. Ignore that for now.
 */
static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
{
	struct request_queue *q = hctx->queue;
	struct request *rq;
	LIST_HEAD(rq_list);
	LIST_HEAD(driver_list);
	struct list_head *dptr;
	int queued;

	WARN_ON(!cpumask_test_cpu(raw_smp_processor_id(), hctx->cpumask));

	if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state)))
		return;

	hctx->run++;

	/*
	 * Touch any software queue that has pending entries.
	 */
	flush_busy_ctxs(hctx, &rq_list);

	/*
	 * If we have previous entries on our dispatch list, grab them
	 * and stuff them at the front for more fair dispatch.
	 */
	if (!list_empty_careful(&hctx->dispatch)) {
		spin_lock(&hctx->lock);
		if (!list_empty(&hctx->dispatch))
			list_splice_init(&hctx->dispatch, &rq_list);
		spin_unlock(&hctx->lock);
	}

	/*
	 * Start off with dptr being NULL, so we start the first request
	 * immediately, even if we have more pending.
	 */
	dptr = NULL;

	/*
	 * Now process all the entries, sending them to the driver.
	 */
	queued = 0;
	while (!list_empty(&rq_list)) {
		struct blk_mq_queue_data bd;
		int ret;

		rq = list_first_entry(&rq_list, struct request, queuelist);
		list_del_init(&rq->queuelist);

		bd.rq = rq;
		bd.list = dptr;
		bd.last = list_empty(&rq_list);

		ret = q->mq_ops->queue_rq(hctx, &bd);
		switch (ret) {
		case BLK_MQ_RQ_QUEUE_OK:
			queued++;
			continue;
		case BLK_MQ_RQ_QUEUE_BUSY:
			list_add(&rq->queuelist, &rq_list);
			__blk_mq_requeue_request(rq);
			break;
		default:
			pr_err("blk-mq: bad return on queue: %d\n", ret);
		case BLK_MQ_RQ_QUEUE_ERROR:
			rq->errors = -EIO;
			blk_mq_end_request(rq, rq->errors);
			break;
		}

		if (ret == BLK_MQ_RQ_QUEUE_BUSY)
			break;

		/*
		 * We've done the first request. If we have more than 1
		 * left in the list, set dptr to defer issue.
		 */
		if (!dptr && rq_list.next != rq_list.prev)
			dptr = &driver_list;
	}

	if (!queued)
		hctx->dispatched[0]++;
	else if (queued < (1 << (BLK_MQ_MAX_DISPATCH_ORDER - 1)))
		hctx->dispatched[ilog2(queued) + 1]++;

	/*
	 * Any items that need requeuing? Stuff them into hctx->dispatch,
	 * that is where we will continue on next queue run.
	 */
	if (!list_empty(&rq_list)) {
		spin_lock(&hctx->lock);
		list_splice(&rq_list, &hctx->dispatch);
		spin_unlock(&hctx->lock);
	}
}

/*
 * It'd be great if the workqueue API had a way to pass
 * in a mask and had some smarts for more clever placement.
 * For now we just round-robin here, switching for every
 * BLK_MQ_CPU_WORK_BATCH queued items.
 */
static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx)
{
	if (hctx->queue->nr_hw_queues == 1)
		return WORK_CPU_UNBOUND;

	if (--hctx->next_cpu_batch <= 0) {
		int cpu = hctx->next_cpu, next_cpu;

		next_cpu = cpumask_next(hctx->next_cpu, hctx->cpumask);
		if (next_cpu >= nr_cpu_ids)
			next_cpu = cpumask_first(hctx->cpumask);

		hctx->next_cpu = next_cpu;
		hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH;

		return cpu;
	}

	return hctx->next_cpu;
}

void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
{
	if (unlikely(test_bit(BLK_MQ_S_STOPPED, &hctx->state) ||
	    !blk_mq_hw_queue_mapped(hctx)))
		return;

	if (!async) {
		int cpu = get_cpu();
		if (cpumask_test_cpu(cpu, hctx->cpumask)) {
			__blk_mq_run_hw_queue(hctx);
			put_cpu();
			return;
		}

		put_cpu();
	}

	kblockd_schedule_delayed_work_on(blk_mq_hctx_next_cpu(hctx),
			&hctx->run_work, 0);
}

void blk_mq_run_queues(struct request_queue *q, bool async)
{
	struct blk_mq_hw_ctx *hctx;
	int i;

	queue_for_each_hw_ctx(q, hctx, i) {
		if ((!blk_mq_hctx_has_pending(hctx) &&
		    list_empty_careful(&hctx->dispatch)) ||
		    test_bit(BLK_MQ_S_STOPPED, &hctx->state))
			continue;

		blk_mq_run_hw_queue(hctx, async);
	}
}
EXPORT_SYMBOL(blk_mq_run_queues);

void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx)
{
	cancel_delayed_work(&hctx->run_work);
	cancel_delayed_work(&hctx->delay_work);
	set_bit(BLK_MQ_S_STOPPED, &hctx->state);
}
EXPORT_SYMBOL(blk_mq_stop_hw_queue);

void blk_mq_stop_hw_queues(struct request_queue *q)
{
	struct blk_mq_hw_ctx *hctx;
	int i;

	queue_for_each_hw_ctx(q, hctx, i)
		blk_mq_stop_hw_queue(hctx);
}
EXPORT_SYMBOL(blk_mq_stop_hw_queues);

void blk_mq_start_hw_queue(struct blk_mq_hw_ctx *hctx)
{
	clear_bit(BLK_MQ_S_STOPPED, &hctx->state);

	blk_mq_run_hw_queue(hctx, false);
}
EXPORT_SYMBOL(blk_mq_start_hw_queue);

void blk_mq_start_hw_queues(struct request_queue *q)
{
	struct blk_mq_hw_ctx *hctx;
	int i;

	queue_for_each_hw_ctx(q, hctx, i)
		blk_mq_start_hw_queue(hctx);
}
EXPORT_SYMBOL(blk_mq_start_hw_queues);


void blk_mq_start_stopped_hw_queues(struct request_queue *q, bool async)
{
	struct blk_mq_hw_ctx *hctx;
	int i;

	queue_for_each_hw_ctx(q, hctx, i) {
		if (!test_bit(BLK_MQ_S_STOPPED, &hctx->state))
			continue;

		clear_bit(BLK_MQ_S_STOPPED, &hctx->state);
		blk_mq_run_hw_queue(hctx, async);
	}
}
EXPORT_SYMBOL(blk_mq_start_stopped_hw_queues);

static void blk_mq_run_work_fn(struct work_struct *work)
{
	struct blk_mq_hw_ctx *hctx;

	hctx = container_of(work, struct blk_mq_hw_ctx, run_work.work);

	__blk_mq_run_hw_queue(hctx);
}

static void blk_mq_delay_work_fn(struct work_struct *work)
{
	struct blk_mq_hw_ctx *hctx;

	hctx = container_of(work, struct blk_mq_hw_ctx, delay_work.work);

	if (test_and_clear_bit(BLK_MQ_S_STOPPED, &hctx->state))
		__blk_mq_run_hw_queue(hctx);
}

void blk_mq_delay_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs)
{
	if (unlikely(!blk_mq_hw_queue_mapped(hctx)))
		return;

	kblockd_schedule_delayed_work_on(blk_mq_hctx_next_cpu(hctx),
			&hctx->delay_work, msecs_to_jiffies(msecs));
}
EXPORT_SYMBOL(blk_mq_delay_queue);

static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx,
				    struct request *rq, bool at_head)
{
	struct blk_mq_ctx *ctx = rq->mq_ctx;

	trace_block_rq_insert(hctx->queue, rq);

	if (at_head)
		list_add(&rq->queuelist, &ctx->rq_list);
	else
		list_add_tail(&rq->queuelist, &ctx->rq_list);

	blk_mq_hctx_mark_pending(hctx, ctx);
}

void blk_mq_insert_request(struct request *rq, bool at_head, bool run_queue,
		bool async)
{
	struct request_queue *q = rq->q;
	struct blk_mq_hw_ctx *hctx;
	struct blk_mq_ctx *ctx = rq->mq_ctx, *current_ctx;

	current_ctx = blk_mq_get_ctx(q);
	if (!cpu_online(ctx->cpu))
		rq->mq_ctx = ctx = current_ctx;

	hctx = q->mq_ops->map_queue(q, ctx->cpu);

	spin_lock(&ctx->lock);
	__blk_mq_insert_request(hctx, rq, at_head);
	spin_unlock(&ctx->lock);

	if (run_queue)
		blk_mq_run_hw_queue(hctx, async);

	blk_mq_put_ctx(current_ctx);
}

static void blk_mq_insert_requests(struct request_queue *q,
				     struct blk_mq_ctx *ctx,
				     struct list_head *list,
				     int depth,
				     bool from_schedule)

{
	struct blk_mq_hw_ctx *hctx;
	struct blk_mq_ctx *current_ctx;

	trace_block_unplug(q, depth, !from_schedule);