Merge branch 'for-3.16/core' into for-3.16/drivers

mtip32xx uses blk_mq_alloc_reserved_request(), so pull in the
core changes so we have a properly merged end result.
Signed-off-by: Jens Axboe <axboe@fb.com>

Documentation/DocBook/filesystems.tmpl

@@ -62,7 +62,7 @@
 !Efs/mpage.c
 !Efs/namei.c
 !Efs/buffer.c
-!Efs/bio.c
+!Eblock/bio.c
 !Efs/seq_file.c
 !Efs/filesystems.c
 !Efs/fs-writeback.c

block/Makefile

@@ -2,13 +2,15 @@
 # Makefile for the kernel block layer
 #
 
-obj-$(CONFIG_BLOCK) := elevator.o blk-core.o blk-tag.o blk-sysfs.o \
+obj-$(CONFIG_BLOCK) := bio.o elevator.o blk-core.o blk-tag.o blk-sysfs.o \
 			blk-flush.o blk-settings.o blk-ioc.o blk-map.o \
 			blk-exec.o blk-merge.o blk-softirq.o blk-timeout.o \
 			blk-iopoll.o blk-lib.o blk-mq.o blk-mq-tag.o \
 			blk-mq-sysfs.o blk-mq-cpu.o blk-mq-cpumap.o ioctl.o \
-			genhd.o scsi_ioctl.o partition-generic.o partitions/
+			genhd.o scsi_ioctl.o partition-generic.o ioprio.o \
+			partitions/
 
+obj-$(CONFIG_BOUNCE)	+= bounce.o
 obj-$(CONFIG_BLK_DEV_BSG)	+= bsg.o
 obj-$(CONFIG_BLK_DEV_BSGLIB)	+= bsg-lib.o
 obj-$(CONFIG_BLK_CGROUP)	+= blk-cgroup.o
@@ -20,3 +22,4 @@ obj-$(CONFIG_IOSCHED_CFQ)	+= cfq-iosched.o
 obj-$(CONFIG_BLOCK_COMPAT)	+= compat_ioctl.o
 obj-$(CONFIG_BLK_DEV_INTEGRITY)	+= blk-integrity.o
 obj-$(CONFIG_BLK_CMDLINE_PARSER)	+= cmdline-parser.o
+obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o

fs/bio-integrity.c → block/bio-integrity.c

@@ -617,7 +617,7 @@ int bioset_integrity_create(struct bio_set *bs, int pool_size)
 	if (!bs->bio_integrity_pool)
 		return -1;
 
-	bs->bvec_integrity_pool = biovec_create_pool(bs, pool_size);
+	bs->bvec_integrity_pool = biovec_create_pool(pool_size);
 	if (!bs->bvec_integrity_pool) {
 		mempool_destroy(bs->bio_integrity_pool);
 		return -1;

fs/bio.c → block/bio.c

@@ -305,6 +305,8 @@ static void bio_chain_endio(struct bio *bio, int error)
 
 /**
  * bio_chain - chain bio completions
+ * @bio: the target bio
+ * @parent: the @bio's parent bio
  *
  * The caller won't have a bi_end_io called when @bio completes - instead,
  * @parent's bi_end_io won't be called until both @parent and @bio have
@@ -1011,8 +1013,7 @@ static void bio_set_map_data(struct bio_map_data *bmd, struct bio *bio,
 	bio->bi_private = bmd;
 }
 
-static struct bio_map_data *bio_alloc_map_data(int nr_segs,
-					       unsigned int iov_count,
+static struct bio_map_data *bio_alloc_map_data(unsigned int iov_count,
 					       gfp_t gfp_mask)
 {
 	if (iov_count > UIO_MAXIOV)
@@ -1154,7 +1155,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
 	if (offset)
 		nr_pages++;
 
-	bmd = bio_alloc_map_data(nr_pages, iov_count, gfp_mask);
+	bmd = bio_alloc_map_data(iov_count, gfp_mask);
 	if (!bmd)
 		return ERR_PTR(-ENOMEM);
 
@@ -1859,7 +1860,7 @@ EXPORT_SYMBOL_GPL(bio_trim);
  * create memory pools for biovec's in a bio_set.
  * use the global biovec slabs created for general use.
  */
-mempool_t *biovec_create_pool(struct bio_set *bs, int pool_entries)
+mempool_t *biovec_create_pool(int pool_entries)
 {
 	struct biovec_slab *bp = bvec_slabs + BIOVEC_MAX_IDX;
 
@@ -1922,7 +1923,7 @@ struct bio_set *bioset_create(unsigned int pool_size, unsigned int front_pad)
 	if (!bs->bio_pool)
 		goto bad;
 
-	bs->bvec_pool = biovec_create_pool(bs, pool_size);
+	bs->bvec_pool = biovec_create_pool(pool_size);
 	if (!bs->bvec_pool)
 		goto bad;
 

block/blk-core.c

@@ -576,12 +576,9 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 	if (!q)
 		return NULL;
 
-	if (percpu_counter_init(&q->mq_usage_counter, 0))
-		goto fail_q;
-
 	q->id = ida_simple_get(&blk_queue_ida, 0, 0, gfp_mask);
 	if (q->id < 0)
-		goto fail_c;
+		goto fail_q;
 
 	q->backing_dev_info.ra_pages =
 			(VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
@@ -639,8 +636,6 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 	bdi_destroy(&q->backing_dev_info);
 fail_id:
 	ida_simple_remove(&blk_queue_ida, q->id);
-fail_c:
-	percpu_counter_destroy(&q->mq_usage_counter);
 fail_q:
 	kmem_cache_free(blk_requestq_cachep, q);
 	return NULL;
@@ -848,6 +843,47 @@ static void freed_request(struct request_list *rl, unsigned int flags)
 		__freed_request(rl, sync ^ 1);
 }
 
+int blk_update_nr_requests(struct request_queue *q, unsigned int nr)
+{
+	struct request_list *rl;
+
+	spin_lock_irq(q->queue_lock);
+	q->nr_requests = nr;
+	blk_queue_congestion_threshold(q);
+
+	/* congestion isn't cgroup aware and follows root blkcg for now */
+	rl = &q->root_rl;
+
+	if (rl->count[BLK_RW_SYNC] >= queue_congestion_on_threshold(q))
+		blk_set_queue_congested(q, BLK_RW_SYNC);
+	else if (rl->count[BLK_RW_SYNC] < queue_congestion_off_threshold(q))
+		blk_clear_queue_congested(q, BLK_RW_SYNC);
+
+	if (rl->count[BLK_RW_ASYNC] >= queue_congestion_on_threshold(q))
+		blk_set_queue_congested(q, BLK_RW_ASYNC);
+	else if (rl->count[BLK_RW_ASYNC] < queue_congestion_off_threshold(q))
+		blk_clear_queue_congested(q, BLK_RW_ASYNC);
+
+	blk_queue_for_each_rl(rl, q) {
+		if (rl->count[BLK_RW_SYNC] >= q->nr_requests) {
+			blk_set_rl_full(rl, BLK_RW_SYNC);
+		} else {
+			blk_clear_rl_full(rl, BLK_RW_SYNC);
+			wake_up(&rl->wait[BLK_RW_SYNC]);
+		}
+
+		if (rl->count[BLK_RW_ASYNC] >= q->nr_requests) {
+			blk_set_rl_full(rl, BLK_RW_ASYNC);
+		} else {
+			blk_clear_rl_full(rl, BLK_RW_ASYNC);
+			wake_up(&rl->wait[BLK_RW_ASYNC]);
+		}
+	}
+
+	spin_unlock_irq(q->queue_lock);
+	return 0;
+}
+
 /*
  * Determine if elevator data should be initialized when allocating the
  * request associated with @bio.
@@ -1137,7 +1173,7 @@ static struct request *blk_old_get_request(struct request_queue *q, int rw,
 struct request *blk_get_request(struct request_queue *q, int rw, gfp_t gfp_mask)
 {
 	if (q->mq_ops)
-		return blk_mq_alloc_request(q, rw, gfp_mask);
+		return blk_mq_alloc_request(q, rw, gfp_mask, false);
 	else
 		return blk_old_get_request(q, rw, gfp_mask);
 }
@@ -1233,12 +1269,15 @@ static void add_acct_request(struct request_queue *q, struct request *rq,
 static void part_round_stats_single(int cpu, struct hd_struct *part,
 				    unsigned long now)
 {
+	int inflight;
+
 	if (now == part->stamp)
 		return;
 
-	if (part_in_flight(part)) {
+	inflight = part_in_flight(part);
+	if (inflight) {
 		__part_stat_add(cpu, part, time_in_queue,
-				part_in_flight(part) * (now - part->stamp));
+				inflight * (now - part->stamp));
 		__part_stat_add(cpu, part, io_ticks, (now - part->stamp));
 	}
 	part->stamp = now;
@@ -1427,6 +1466,8 @@ bool bio_attempt_front_merge(struct request_queue *q, struct request *req,
  * added on the elevator at this point.  In addition, we don't have
  * reliable access to the elevator outside queue lock.  Only check basic
  * merging parameters without querying the elevator.
+ *
+ * Caller must ensure !blk_queue_nomerges(q) beforehand.
  */
 bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio,
 			    unsigned int *request_count)
@@ -1436,9 +1477,6 @@ bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio,
 	bool ret = false;
 	struct list_head *plug_list;
 
-	if (blk_queue_nomerges(q))
-		goto out;
-
 	plug = current->plug;
 	if (!plug)
 		goto out;
@@ -1517,7 +1555,8 @@ void blk_queue_bio(struct request_queue *q, struct bio *bio)
 	 * Check if we can merge with the plugged list before grabbing
 	 * any locks.
 	 */
-	if (blk_attempt_plug_merge(q, bio, &request_count))
+	if (!blk_queue_nomerges(q) &&
+	    blk_attempt_plug_merge(q, bio, &request_count))
 		return;
 
 	spin_lock_irq(q->queue_lock);

block/blk-flush.c

@@ -130,21 +130,13 @@ static void blk_flush_restore_request(struct request *rq)
 	blk_clear_rq_complete(rq);
 }
 
-static void mq_flush_run(struct work_struct *work)
-{
-	struct request *rq;
-
-	rq = container_of(work, struct request, requeue_work);
-
-	memset(&rq->csd, 0, sizeof(rq->csd));
-	blk_mq_insert_request(rq, false, true, false);
-}
-
 static bool blk_flush_queue_rq(struct request *rq, bool add_front)
 {
 	if (rq->q->mq_ops) {
-		INIT_WORK(&rq->requeue_work, mq_flush_run);
-		kblockd_schedule_work(&rq->requeue_work);
+		struct request_queue *q = rq->q;
+
+		blk_mq_add_to_requeue_list(rq, add_front);
+		blk_mq_kick_requeue_list(q);
 		return false;
 	} else {
 		if (add_front)

block/blk-iopoll.c

@@ -64,12 +64,12 @@ EXPORT_SYMBOL(__blk_iopoll_complete);
  *     iopoll handler will not be invoked again before blk_iopoll_sched_prep()
  *     is called.
  **/
-void blk_iopoll_complete(struct blk_iopoll *iopoll)
+void blk_iopoll_complete(struct blk_iopoll *iop)
 {
 	unsigned long flags;
 
 	local_irq_save(flags);
-	__blk_iopoll_complete(iopoll);
+	__blk_iopoll_complete(iop);
 	local_irq_restore(flags);
 }
 EXPORT_SYMBOL(blk_iopoll_complete);

block/blk-lib.c

@@ -226,7 +226,7 @@ EXPORT_SYMBOL(blkdev_issue_write_same);
  *  Generate and issue number of bios with zerofiled pages.
  */
 
-int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
+static int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 				  sector_t nr_sects, gfp_t gfp_mask)
 {
 	int ret;

block/blk-mq-cpu.c

@@ -18,14 +18,18 @@ static int blk_mq_main_cpu_notify(struct notifier_block *self,
 {
 	unsigned int cpu = (unsigned long) hcpu;
 	struct blk_mq_cpu_notifier *notify;
+	int ret = NOTIFY_OK;
 
 	raw_spin_lock(&blk_mq_cpu_notify_lock);
 
-	list_for_each_entry(notify, &blk_mq_cpu_notify_list, list)
-		notify->notify(notify->data, action, cpu);
+	list_for_each_entry(notify, &blk_mq_cpu_notify_list, list) {
+		ret = notify->notify(notify->data, action, cpu);
+		if (ret != NOTIFY_OK)
+			break;
+	}
 
 	raw_spin_unlock(&blk_mq_cpu_notify_lock);
-	return NOTIFY_OK;
+	return ret;
 }
 
 void blk_mq_register_cpu_notifier(struct blk_mq_cpu_notifier *notifier)
@@ -45,7 +49,7 @@ void blk_mq_unregister_cpu_notifier(struct blk_mq_cpu_notifier *notifier)
 }
 
 void blk_mq_init_cpu_notifier(struct blk_mq_cpu_notifier *notifier,
-			      void (*fn)(void *, unsigned long, unsigned int),
+			      int (*fn)(void *, unsigned long, unsigned int),
 			      void *data)
 {
 	notifier->notify = fn;

block/blk-mq-cpumap.c

@@ -96,3 +96,19 @@ unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set)
 	kfree(map);
 	return NULL;
 }
+
+/*
+ * We have no quick way of doing reverse lookups. This is only used at
+ * queue init time, so runtime isn't important.
+ */
+int blk_mq_hw_queue_to_node(unsigned int *mq_map, unsigned int index)
+{
+	int i;
+
+	for_each_possible_cpu(i) {
+		if (index == mq_map[i])
+			return cpu_to_node(i);
+	}
+
+	return NUMA_NO_NODE;
+}

block/blk-mq-sysfs.c

@@ -203,45 +203,14 @@ static ssize_t blk_mq_hw_sysfs_rq_list_show(struct blk_mq_hw_ctx *hctx,
 	return ret;
 }
 
-static ssize_t blk_mq_hw_sysfs_ipi_show(struct blk_mq_hw_ctx *hctx, char *page)
-{
-	ssize_t ret;
-
-	spin_lock(&hctx->lock);
-	ret = sprintf(page, "%u\n", !!(hctx->flags & BLK_MQ_F_SHOULD_IPI));
-	spin_unlock(&hctx->lock);
-
-	return ret;
-}
-
-static ssize_t blk_mq_hw_sysfs_ipi_store(struct blk_mq_hw_ctx *hctx,
-					 const char *page, size_t len)
+static ssize_t blk_mq_hw_sysfs_tags_show(struct blk_mq_hw_ctx *hctx, char *page)
 {
-	struct blk_mq_ctx *ctx;
-	unsigned long ret;
-	unsigned int i;
-
-	if (kstrtoul(page, 10, &ret)) {
-		pr_err("blk-mq-sysfs: invalid input '%s'\n", page);
-		return -EINVAL;
-	}
-
-	spin_lock(&hctx->lock);
-	if (ret)
-		hctx->flags |= BLK_MQ_F_SHOULD_IPI;
-	else
-		hctx->flags &= ~BLK_MQ_F_SHOULD_IPI;
-	spin_unlock(&hctx->lock);
-
-	hctx_for_each_ctx(hctx, ctx, i)
-		ctx->ipi_redirect = !!ret;
-
-	return len;
+	return blk_mq_tag_sysfs_show(hctx->tags, page);
 }
 
-static ssize_t blk_mq_hw_sysfs_tags_show(struct blk_mq_hw_ctx *hctx, char *page)
+static ssize_t blk_mq_hw_sysfs_active_show(struct blk_mq_hw_ctx *hctx, char *page)
 {
-	return blk_mq_tag_sysfs_show(hctx->tags, page);
+	return sprintf(page, "%u\n", atomic_read(&hctx->nr_active));
 }
 
 static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page)
@@ -303,15 +272,14 @@ static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_dispatched = {
 	.attr = {.name = "dispatched", .mode = S_IRUGO },
 	.show = blk_mq_hw_sysfs_dispatched_show,
 };
+static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_active = {
+	.attr = {.name = "active", .mode = S_IRUGO },
+	.show = blk_mq_hw_sysfs_active_show,
+};
 static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_pending = {
 	.attr = {.name = "pending", .mode = S_IRUGO },
 	.show = blk_mq_hw_sysfs_rq_list_show,
 };
-static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_ipi = {
-	.attr = {.name = "ipi_redirect", .mode = S_IRUGO | S_IWUSR},
-	.show = blk_mq_hw_sysfs_ipi_show,
-	.store = blk_mq_hw_sysfs_ipi_store,
-};
 static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_tags = {
 	.attr = {.name = "tags", .mode = S_IRUGO },
 	.show = blk_mq_hw_sysfs_tags_show,
@@ -326,9 +294,9 @@ static struct attribute *default_hw_ctx_attrs[] = {
 	&blk_mq_hw_sysfs_run.attr,
 	&blk_mq_hw_sysfs_dispatched.attr,
 	&blk_mq_hw_sysfs_pending.attr,
-	&blk_mq_hw_sysfs_ipi.attr,
 	&blk_mq_hw_sysfs_tags.attr,
 	&blk_mq_hw_sysfs_cpus.attr,
+	&blk_mq_hw_sysfs_active.attr,
 	NULL,
 };
 

block/blk-mq-tag.h

 #ifndef INT_BLK_MQ_TAG_H
 #define INT_BLK_MQ_TAG_H
 
-#include <linux/percpu_ida.h>
+#include "blk-mq.h"
+
+enum {
+	BT_WAIT_QUEUES	= 8,
+	BT_WAIT_BATCH	= 8,
+};
+
+struct bt_wait_state {
+	atomic_t wait_cnt;
+	wait_queue_head_t wait;
+} ____cacheline_aligned_in_smp;
+
+#define TAG_TO_INDEX(bt, tag)	((tag) >> (bt)->bits_per_word)
+#define TAG_TO_BIT(bt, tag)	((tag) & ((1 << (bt)->bits_per_word) - 1))
+
+struct blk_mq_bitmap_tags {
+	unsigned int depth;
+	unsigned int wake_cnt;
+	unsigned int bits_per_word;
+
+	unsigned int map_nr;
+	struct blk_align_bitmap *map;
+
+	unsigned int wake_index;
+	struct bt_wait_state *bs;
+};
 
 /*
  * Tag address space map.
@@ -9,11 +34,11 @@
 struct blk_mq_tags {
 	unsigned int nr_tags;
 	unsigned int nr_reserved_tags;
-	unsigned int nr_batch_move;
-	unsigned int nr_max_cache;
 
-	struct percpu_ida free_tags;
-	struct percpu_ida reserved_tags;
+	atomic_t active_queues;
+
+	struct blk_mq_bitmap_tags bitmap_tags;
+	struct blk_mq_bitmap_tags breserved_tags;
 
 	struct request **rqs;
 	struct list_head page_list;
@@ -23,12 +48,12 @@ struct blk_mq_tags {
 extern struct blk_mq_tags *blk_mq_init_tags(unsigned int nr_tags, unsigned int reserved_tags, int node);
 extern void blk_mq_free_tags(struct blk_mq_tags *tags);
 
-extern unsigned int blk_mq_get_tag(struct blk_mq_tags *tags, gfp_t gfp, bool reserved);
-extern void blk_mq_wait_for_tags(struct blk_mq_tags *tags);
-extern void blk_mq_put_tag(struct blk_mq_tags *tags, unsigned int tag);
-extern void blk_mq_tag_busy_iter(struct blk_mq_tags *tags, void (*fn)(void *data, unsigned long *), void *data);
+extern unsigned int blk_mq_get_tag(struct blk_mq_hw_ctx *hctx, unsigned int *last_tag, gfp_t gfp, bool reserved);
+extern void blk_mq_put_tag(struct blk_mq_hw_ctx *hctx, unsigned int tag, unsigned int *last_tag);
 extern bool blk_mq_has_free_tags(struct blk_mq_tags *tags);
 extern ssize_t blk_mq_tag_sysfs_show(struct blk_mq_tags *tags, char *page);
+extern void blk_mq_tag_init_last_tag(struct blk_mq_tags *tags, unsigned int *last_tag);
+extern int blk_mq_tag_update_depth(struct blk_mq_tags *tags, unsigned int depth);
 
 enum {
 	BLK_MQ_TAG_CACHE_MIN	= 1,
@@ -41,4 +66,23 @@ enum {
 	BLK_MQ_TAG_MAX		= BLK_MQ_TAG_FAIL - 1,
 };
 
+extern bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *);
+extern void __blk_mq_tag_idle(struct blk_mq_hw_ctx *);
+
+static inline bool blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx)
+{
+	if (!(hctx->flags & BLK_MQ_F_TAG_SHARED))
+		return false;
+
+	return __blk_mq_tag_busy(hctx);
+}
+
+static inline void blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx)
+{
+	if (!(hctx->flags & BLK_MQ_F_TAG_SHARED))
+		return;
+
+	__blk_mq_tag_idle(hctx);
+}
+
 #endif

block/blk-mq.h

@@ -11,7 +11,8 @@ struct blk_mq_ctx {
 
 	unsigned int		cpu;
 	unsigned int		index_hw;
-	unsigned int		ipi_redirect;
+
+	unsigned int		last_tag ____cacheline_aligned_in_smp;
 
 	/* incremented at dispatch time */
 	unsigned long		rq_dispatched[2];
@@ -22,7 +23,7 @@ struct blk_mq_ctx {
 
 	struct request_queue	*queue;
 	struct kobject		kobj;
-};
+} ____cacheline_aligned_in_smp;
 
 void __blk_mq_complete_request(struct request *rq);
 void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async);
@@ -31,13 +32,14 @@ void blk_mq_drain_queue(struct request_queue *q);
 void blk_mq_free_queue(struct request_queue *q);
 void blk_mq_clone_flush_request(struct request *flush_rq,
 		struct request *orig_rq);
+int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr);
 
 /*
  * CPU hotplug helpers
  */
 struct blk_mq_cpu_notifier;
 void blk_mq_init_cpu_notifier(struct blk_mq_cpu_notifier *notifier,
-			      void (*fn)(void *, unsigned long, unsigned int),
+			      int (*fn)(void *, unsigned long, unsigned int),
 			      void *data);
 void blk_mq_register_cpu_notifier(struct blk_mq_cpu_notifier *notifier);
 void blk_mq_unregister_cpu_notifier(struct blk_mq_cpu_notifier *notifier);
@@ -50,7 +52,15 @@ void blk_mq_disable_hotplug(void);
  */
 extern unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set);
 extern int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues);
+extern int blk_mq_hw_queue_to_node(unsigned int *map, unsigned int);
 
-void blk_mq_add_timer(struct request *rq);
+/*
+ * Basic implementation of sparser bitmap, allowing the user to spread
+ * the bits over more cachelines.
+ */
+struct blk_align_bitmap {
+	unsigned long word;
+	unsigned long depth;
+} ____cacheline_aligned_in_smp;
 
 #endif

block/blk-sysfs.c

@@ -48,11 +48,10 @@ static ssize_t queue_requests_show(struct request_queue *q, char *page)
 static ssize_t
 queue_requests_store(struct request_queue *q, const char *page, size_t count)
 {
-	struct request_list *rl;
 	unsigned long nr;
-	int ret;
+	int ret, err;
 
-	if (!q->request_fn)
+	if (!q->request_fn && !q->mq_ops)
 		return -EINVAL;
 
 	ret = queue_var_store(&nr, page, count);
@@ -62,40 +61,14 @@ queue_requests_store(struct request_queue *q, const char *page, size_t count)
 	if (nr < BLKDEV_MIN_RQ)
 		nr = BLKDEV_MIN_RQ;
 
-	spin_lock_irq(q->queue_lock);
-	q->nr_requests = nr;
-	blk_queue_congestion_threshold(q);
-
-	/* congestion isn't cgroup aware and follows root blkcg for now */
-	rl = &q->root_rl;
-
-	if (rl->count[BLK_RW_SYNC] >= queue_congestion_on_threshold(q))
-		blk_set_queue_congested(q, BLK_RW_SYNC);
-	else if (rl->count[BLK_RW_SYNC] < queue_congestion_off_threshold(q))
-		blk_clear_queue_congested(q, BLK_RW_SYNC);
-
-	if (rl->count[BLK_RW_ASYNC] >= queue_congestion_on_threshold(q))
-		blk_set_queue_congested(q, BLK_RW_ASYNC);
-	else if (rl->count[BLK_RW_ASYNC] < queue_congestion_off_threshold(q))
-		blk_clear_queue_congested(q, BLK_RW_ASYNC);
-
-	blk_queue_for_each_rl(rl, q) {
-		if (rl->count[BLK_RW_SYNC] >= q->nr_requests) {
-			blk_set_rl_full(rl, BLK_RW_SYNC);
-		} else {
-			blk_clear_rl_full(rl, BLK_RW_SYNC);
-			wake_up(&rl->wait[BLK_RW_SYNC]);
-		}
+	if (q->request_fn)
+		err = blk_update_nr_requests(q, nr);
+	else
+		err = blk_mq_update_nr_requests(q, nr);
 
-		if (rl->count[BLK_RW_ASYNC] >= q->nr_requests) {
-			blk_set_rl_full(rl, BLK_RW_ASYNC);
-		} else {
-			blk_clear_rl_full(rl, BLK_RW_ASYNC);
-			wake_up(&rl->wait[BLK_RW_ASYNC]);
-		}
-	}
+	if (err)
+		return err;
 
-	spin_unlock_irq(q->queue_lock);
 	return ret;
 }
 
@@ -544,8 +517,6 @@ static void blk_release_queue(struct kobject *kobj)
 	if (q->queue_tags)
 		__blk_queue_free_tags(q);
 
-	percpu_counter_destroy(&q->mq_usage_counter);
-
 	if (q->mq_ops)
 		blk_mq_free_queue(q);
 

block/blk-throttle.c

@@ -744,7 +744,7 @@ static inline void throtl_extend_slice(struct throtl_grp *tg, bool rw,
 static bool throtl_slice_used(struct throtl_grp *tg, bool rw)
 {
 	if (time_in_range(jiffies, tg->slice_start[rw], tg->slice_end[rw]))
-		return 0;
+		return false;
 
 	return 1;
 }
@@ -842,7 +842,7 @@ static bool tg_with_in_iops_limit(struct throtl_grp *tg, struct bio *bio,
 	if (tg->io_disp[rw] + 1 <= io_allowed) {
 		if (wait)
 			*wait = 0;
-		return 1;
+		return true;
 	}
 
 	/* Calc approx time to dispatch */
@@ -880,7 +880,7 @@ static bool tg_with_in_bps_limit(struct throtl_grp *tg, struct bio *bio,
 	if (tg->bytes_disp[rw] + bio->bi_iter.bi_size <= bytes_allowed) {
 		if (wait)
 			*wait = 0;
-		return 1;
+		return true;
 	}
 
 	/* Calc approx time to dispatch */
@@ -923,7 +923,7 @@ static bool tg_may_dispatch(struct throtl_grp *tg, struct bio *bio,
 	if (tg->bps[rw] == -1 && tg->iops[rw] == -1) {
 		if (wait)
 			*wait = 0;
-		return 1;
+		return true;
 	}
 
 	/*
@@ -1258,7 +1258,7 @@ static void throtl_pending_timer_fn(unsigned long arg)
  * of throtl_data->service_queue.  Those bio's are ready and issued by this
  * function.
  */
-void blk_throtl_dispatch_work_fn(struct work_struct *work)
+static void blk_throtl_dispatch_work_fn(struct work_struct *work)
 {
 	struct throtl_data *td = container_of(work, struct throtl_data,
 					      dispatch_work);

block/blk-timeout.c

@@ -96,11 +96,7 @@ static void blk_rq_timed_out(struct request *req)
 			__blk_complete_request(req);
 		break;
 	case BLK_EH_RESET_TIMER:
-		if (q->mq_ops)
-			blk_mq_add_timer(req);
-		else
 		blk_add_timer(req);
-
 		blk_clear_rq_complete(req);
 		break;
 	case BLK_EH_NOT_HANDLED:
@@ -170,7 +166,26 @@ void blk_abort_request(struct request *req)
 }
 EXPORT_SYMBOL_GPL(blk_abort_request);
 
-void __blk_add_timer(struct request *req, struct list_head *timeout_list)
+unsigned long blk_rq_timeout(unsigned long timeout)
+{
+	unsigned long maxt;
+
+	maxt = round_jiffies_up(jiffies + BLK_MAX_TIMEOUT);
+	if (time_after(timeout, maxt))
+		timeout = maxt;
+
+	return timeout;
+}
+
+/**
+ * blk_add_timer - Start timeout timer for a single request
+ * @req:	request that is about to start running.
+ *
+ * Notes:
+ *    Each request has its own timer, and as it is added to the queue, we
+ *    set up the timer. When the request completes, we cancel the timer.
+ */
+void blk_add_timer(struct request *req)
 {
 	struct request_queue *q = req->q;
 	unsigned long expiry;
@@ -188,15 +203,15 @@ void __blk_add_timer(struct request *req, struct list_head *timeout_list)
 		req->timeout = q->rq_timeout;
 
 	req->deadline = jiffies + req->timeout;
-	if (timeout_list)
-		list_add_tail(&req->timeout_list, timeout_list);
+	if (!q->mq_ops)
+		list_add_tail(&req->timeout_list, &req->q->timeout_list);
 
 	/*
 	 * If the timer isn't already pending or this timeout is earlier
 	 * than an existing one, modify the timer. Round up to next nearest
 	 * second.
 	 */
-	expiry = round_jiffies_up(req->deadline);
+	expiry = blk_rq_timeout(round_jiffies_up(req->deadline));
 
 	if (!timer_pending(&q->timeout) ||
 	    time_before(expiry, q->timeout.expires)) {
@@ -214,17 +229,3 @@ void __blk_add_timer(struct request *req, struct list_head *timeout_list)
 	}
 
 }
-
-/**
- * blk_add_timer - Start timeout timer for a single request
- * @req:	request that is about to start running.
- *
- * Notes:
- *    Each request has its own timer, and as it is added to the queue, we
- *    set up the timer. When the request completes, we cancel the timer.
- */
-void blk_add_timer(struct request *req)
-{
-	__blk_add_timer(req, &req->q->timeout_list);
-}
-

block/blk.h

@@ -9,6 +9,9 @@
 /* Number of requests a "batching" process may submit */
 #define BLK_BATCH_REQ	32
 
+/* Max future timer expiry for timeouts */
+#define BLK_MAX_TIMEOUT		(5 * HZ)
+
 extern struct kmem_cache *blk_requestq_cachep;
 extern struct kmem_cache *request_cachep;
 extern struct kobj_type blk_queue_ktype;
@@ -37,9 +40,9 @@ bool __blk_end_bidi_request(struct request *rq, int error,
 void blk_rq_timed_out_timer(unsigned long data);
 void blk_rq_check_expired(struct request *rq, unsigned long *next_timeout,
 			  unsigned int *next_set);
-void __blk_add_timer(struct request *req, struct list_head *timeout_list);
+unsigned long blk_rq_timeout(unsigned long timeout);
+void blk_add_timer(struct request *req);
 void blk_delete_timer(struct request *);
-void blk_add_timer(struct request *);
 
 
 bool bio_attempt_front_merge(struct request_queue *q, struct request *req,
@@ -185,6 +188,8 @@ static inline int queue_congestion_off_threshold(struct request_queue *q)
 	return q->nr_congestion_off;
 }
 
+extern int blk_update_nr_requests(struct request_queue *, unsigned int);
+
 /*
  * Contribute to IO statistics IFF:
  *

block/cfq-iosched.c

@@ -4460,7 +4460,7 @@ static int cfq_init_queue(struct request_queue *q, struct elevator_type *e)
 static ssize_t
 cfq_var_show(unsigned int var, char *page)
 {
-	return sprintf(page, "%d\n", var);
+	return sprintf(page, "%u\n", var);
 }
 
 static ssize_t

drivers/block/mtip32xx/mtip32xx.c

@@ -178,7 +178,7 @@ static struct mtip_cmd *mtip_get_int_command(struct driver_data *dd)
 {
 	struct request *rq;
 
-	rq = blk_mq_alloc_reserved_request(dd->queue, 0, __GFP_WAIT);
+	rq = blk_mq_alloc_request(dd->queue, 0, __GFP_WAIT, true);
 	return blk_mq_rq_to_pdu(rq);
 }
 

drivers/block/null_blk.c

@@ -322,39 +322,10 @@ static int null_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq)
 }
 
 static struct blk_mq_hw_ctx *null_alloc_hctx(struct blk_mq_tag_set *set,
-		unsigned int hctx_index)
+					     unsigned int hctx_index,
+					     int node)
 {
-	int b_size = DIV_ROUND_UP(set->nr_hw_queues, nr_online_nodes);
-	int tip = (set->nr_hw_queues % nr_online_nodes);
-	int node = 0, i, n;
-
-	/*
-	 * Split submit queues evenly wrt to the number of nodes. If uneven,
-	 * fill the first buckets with one extra, until the rest is filled with
-	 * no extra.
-	 */
-	for (i = 0, n = 1; i < hctx_index; i++, n++) {
-		if (n % b_size == 0) {
-			n = 0;
-			node++;
-
-			tip--;
-			if (!tip)
-				b_size = set->nr_hw_queues / nr_online_nodes;
-		}
-	}
-
-	/*
-	 * A node might not be online, therefore map the relative node id to the
-	 * real node id.
-	 */
-	for_each_online_node(n) {
-		if (!node)
-			break;
-		node--;
-	}
-
-	return kzalloc_node(sizeof(struct blk_mq_hw_ctx), GFP_KERNEL, n);
+	return kzalloc_node(sizeof(struct blk_mq_hw_ctx), GFP_KERNEL, node);
 }
 
 static void null_free_hctx(struct blk_mq_hw_ctx *hctx, unsigned int hctx_index)

drivers/block/skd_main.c

@@ -743,6 +743,7 @@ static void skd_request_fn(struct request_queue *q)
 				break;
 			}
 			skreq->discard_page = 1;
+			req->completion_data = page;
 			skd_prep_discard_cdb(scsi_req, skreq, page, lba, count);
 
 		} else if (flush == SKD_FLUSH_ZERO_SIZE_FIRST) {
@@ -855,10 +856,9 @@ static void skd_end_request(struct skd_device *skdev,
 
 	if ((io_flags & REQ_DISCARD) &&
 		(skreq->discard_page == 1)) {
-		struct bio *bio = req->bio;
 		pr_debug("%s:%s:%d, free the page!",
 			 skdev->name, __func__, __LINE__);
-		__free_page(bio->bi_io_vec->bv_page);
+		__free_page(req->completion_data);
 	}
 
 	if (unlikely(error)) {

drivers/char/random.c

@@ -902,6 +902,7 @@ void add_disk_randomness(struct gendisk *disk)
 	add_timer_randomness(disk->random, 0x100 + disk_devt(disk));
 	trace_add_disk_randomness(disk_devt(disk), ENTROPY_BITS(&input_pool));
 }
+EXPORT_SYMBOL_GPL(add_disk_randomness);
 #endif
 
 /*********************************************************************

drivers/scsi/sd.c

@@ -737,6 +737,7 @@ static int sd_setup_discard_cmnd(struct scsi_device *sdp, struct request *rq)
 		goto out;
 	}
 
+	rq->completion_data = page;
 	blk_add_request_payload(rq, page, len);
 	ret = scsi_setup_blk_pc_cmnd(sdp, rq);
 	rq->__data_len = nr_bytes;
@@ -839,11 +840,9 @@ static void sd_unprep_fn(struct request_queue *q, struct request *rq)
 {
 	struct scsi_cmnd *SCpnt = rq->special;
 
-	if (rq->cmd_flags & REQ_DISCARD) {
-		struct bio *bio = rq->bio;
+	if (rq->cmd_flags & REQ_DISCARD)
+		__free_page(rq->completion_data);
 
-		__free_page(bio->bi_io_vec->bv_page);
-	}
 	if (SCpnt->cmnd != rq->cmd) {
 		mempool_free(SCpnt->cmnd, sd_cdb_pool);
 		SCpnt->cmnd = NULL;

fs/Makefile

@@ -14,14 +14,13 @@ obj-y :=	open.o read_write.o file_table.o super.o \
 		stack.o fs_struct.o statfs.o
 
 ifeq ($(CONFIG_BLOCK),y)
-obj-y +=	buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o
+obj-y +=	buffer.o block_dev.o direct-io.o mpage.o
 else
 obj-y +=	no-block.o
 endif
 
 obj-$(CONFIG_PROC_FS) += proc_namespace.o
 
-obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o
 obj-y				+= notify/
 obj-$(CONFIG_EPOLL)		+= eventpoll.o
 obj-$(CONFIG_ANON_INODES)	+= anon_inodes.o

include/linux/bio.h

@@ -333,7 +333,7 @@ static inline struct bio *bio_next_split(struct bio *bio, int sectors,
 
 extern struct bio_set *bioset_create(unsigned int, unsigned int);
 extern void bioset_free(struct bio_set *);
-extern mempool_t *biovec_create_pool(struct bio_set *bs, int pool_entries);
+extern mempool_t *biovec_create_pool(int pool_entries);
 
 extern struct bio *bio_alloc_bioset(gfp_t, int, struct bio_set *);
 extern void bio_put(struct bio *);

include/linux/blk-mq.h

@@ -8,7 +8,13 @@ struct blk_mq_tags;
 struct blk_mq_cpu_notifier {
 	struct list_head list;
 	void *data;
-	void (*notify)(void *data, unsigned long action, unsigned int cpu);
+	int (*notify)(void *data, unsigned long action, unsigned int cpu);
+};
+
+struct blk_mq_ctxmap {
+	unsigned int map_size;
+	unsigned int bits_per_word;
+	struct blk_align_bitmap *map;
 };
 
 struct blk_mq_hw_ctx {
@@ -21,6 +27,8 @@ struct blk_mq_hw_ctx {
 	struct delayed_work	run_work;
 	struct delayed_work	delay_work;
 	cpumask_var_t		cpumask;
+	int			next_cpu;
+	int			next_cpu_batch;
 
 	unsigned long		flags;		/* BLK_MQ_F_* flags */
 
@@ -29,10 +37,12 @@ struct blk_mq_hw_ctx {
 
 	void			*driver_data;
 
+	struct blk_mq_ctxmap	ctx_map;
+
 	unsigned int		nr_ctx;
 	struct blk_mq_ctx	**ctxs;
-	unsigned int 		nr_ctx_map;
-	unsigned long		*ctx_map;
+
+	unsigned int		wait_index;
 
 	struct blk_mq_tags	*tags;
 
@@ -44,6 +54,8 @@ struct blk_mq_hw_ctx {
 	unsigned int		numa_node;
 	unsigned int		cmd_size;	/* per-request extra data */
 
+	atomic_t		nr_active;
+
 	struct blk_mq_cpu_notifier	cpu_notifier;
 	struct kobject		kobj;
 };
@@ -51,7 +63,7 @@ struct blk_mq_hw_ctx {
 struct blk_mq_tag_set {
 	struct blk_mq_ops	*ops;
 	unsigned int		nr_hw_queues;
-	unsigned int		queue_depth;
+	unsigned int		queue_depth;	/* max hw supported */
 	unsigned int		reserved_tags;
 	unsigned int		cmd_size;	/* per-request extra data */
 	int			numa_node;
@@ -60,12 +72,15 @@ struct blk_mq_tag_set {
 	void			*driver_data;
 
 	struct blk_mq_tags	**tags;
+
+	struct mutex		tag_list_lock;
+	struct list_head	tag_list;
 };
 
 typedef int (queue_rq_fn)(struct blk_mq_hw_ctx *, struct request *);
 typedef struct blk_mq_hw_ctx *(map_queue_fn)(struct request_queue *, const int);
 typedef struct blk_mq_hw_ctx *(alloc_hctx_fn)(struct blk_mq_tag_set *,
-		unsigned int);
+		unsigned int, int);
 typedef void (free_hctx_fn)(struct blk_mq_hw_ctx *, unsigned int);
 typedef int (init_hctx_fn)(struct blk_mq_hw_ctx *, void *, unsigned int);
 typedef void (exit_hctx_fn)(struct blk_mq_hw_ctx *, unsigned int);
@@ -122,11 +137,14 @@ enum {
 
 	BLK_MQ_F_SHOULD_MERGE	= 1 << 0,
 	BLK_MQ_F_SHOULD_SORT	= 1 << 1,
-	BLK_MQ_F_SHOULD_IPI	= 1 << 2,
+	BLK_MQ_F_TAG_SHARED	= 1 << 2,
 
 	BLK_MQ_S_STOPPED	= 0,
+	BLK_MQ_S_TAG_ACTIVE	= 1,
 
 	BLK_MQ_MAX_DEPTH	= 2048,
+
+	BLK_MQ_CPU_WORK_BATCH	= 8,
 };
 
 struct request_queue *blk_mq_init_queue(struct blk_mq_tag_set *);
@@ -142,19 +160,20 @@ void blk_mq_insert_request(struct request *, bool, bool, bool);
 void blk_mq_run_queues(struct request_queue *q, bool async);
 void blk_mq_free_request(struct request *rq);
 bool blk_mq_can_queue(struct blk_mq_hw_ctx *);
-struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp);
-struct request *blk_mq_alloc_reserved_request(struct request_queue *q, int rw, gfp_t gfp);
+struct request *blk_mq_alloc_request(struct request_queue *q, int rw,
+		gfp_t gfp, bool reserved);
 struct request *blk_mq_tag_to_rq(struct blk_mq_tags *tags, unsigned int tag);
 
 struct blk_mq_hw_ctx *blk_mq_map_queue(struct request_queue *, const int ctx_index);
-struct blk_mq_hw_ctx *blk_mq_alloc_single_hw_queue(struct blk_mq_tag_set *, unsigned int);
+struct blk_mq_hw_ctx *blk_mq_alloc_single_hw_queue(struct blk_mq_tag_set *, unsigned int, int);
 void blk_mq_free_single_hw_queue(struct blk_mq_hw_ctx *, unsigned int);
 
 void blk_mq_end_io(struct request *rq, int error);
 void __blk_mq_end_io(struct request *rq, int error);
 
 void blk_mq_requeue_request(struct request *rq);
-
+void blk_mq_add_to_requeue_list(struct request *rq, bool at_head);
+void blk_mq_kick_requeue_list(struct request_queue *q);
 void blk_mq_complete_request(struct request *rq);
 
 void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx);
@@ -163,6 +182,7 @@ void blk_mq_stop_hw_queues(struct request_queue *q);
 void blk_mq_start_hw_queues(struct request_queue *q);
 void blk_mq_start_stopped_hw_queues(struct request_queue *q, bool async);
 void blk_mq_delay_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs);
+void blk_mq_tag_busy_iter(struct blk_mq_tags *tags, void (*fn)(void *data, unsigned long *), void *data);
 
 /*
  * Driver command data is immediately after the request. So subtract request

include/linux/blk_types.h

@@ -190,6 +190,7 @@ enum rq_flag_bits {
 	__REQ_PM,		/* runtime pm request */
 	__REQ_END,		/* last of chain of requests */
 	__REQ_HASHED,		/* on IO scheduler merge hash */
+	__REQ_MQ_INFLIGHT,	/* track inflight for MQ */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -243,5 +244,6 @@ enum rq_flag_bits {
 #define REQ_PM			(1ULL << __REQ_PM)
 #define REQ_END			(1ULL << __REQ_END)
 #define REQ_HASHED		(1ULL << __REQ_HASHED)
+#define REQ_MQ_INFLIGHT		(1ULL << __REQ_MQ_INFLIGHT)
 
 #endif /* __LINUX_BLK_TYPES_H */

include/linux/blkdev.h

@@ -90,15 +90,15 @@ enum rq_cmd_type_bits {
 #define BLK_MAX_CDB	16
 
 /*
- * try to put the fields that are referenced together in the same cacheline.
- * if you modify this structure, be sure to check block/blk-core.c:blk_rq_init()
- * as well!
+ * Try to put the fields that are referenced together in the same cacheline.
+ *
+ * If you modify this structure, make sure to update blk_rq_init() and
+ * especially blk_mq_rq_ctx_init() to take care of the added fields.
  */
 struct request {
 	struct list_head queuelist;
 	union {
 		struct call_single_data csd;
-		struct work_struct requeue_work;
 		unsigned long fifo_time;
 	};
 
@@ -462,6 +462,10 @@ struct request_queue {
 	struct request		*flush_rq;
 	spinlock_t		mq_flush_lock;
 
+	struct list_head	requeue_list;
+	spinlock_t		requeue_lock;
+	struct work_struct	requeue_work;
+
 	struct mutex		sysfs_lock;
 
 	int			bypass_depth;
@@ -480,6 +484,9 @@ struct request_queue {
 	wait_queue_head_t	mq_freeze_wq;
 	struct percpu_counter	mq_usage_counter;
 	struct list_head	all_q_node;
+
+	struct blk_mq_tag_set	*tag_set;
+	struct list_head	tag_set_list;
 };
 
 #define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */

mm/Makefile

@@ -30,7 +30,6 @@ endif
 
 obj-$(CONFIG_HAVE_MEMBLOCK) += memblock.o
 
-obj-$(CONFIG_BOUNCE)	+= bounce.o
 obj-$(CONFIG_SWAP)	+= page_io.o swap_state.o swapfile.o
 obj-$(CONFIG_FRONTSWAP)	+= frontswap.o
 obj-$(CONFIG_ZSWAP)	+= zswap.o