Merge branch 'block-5.7' into for-5.8/block

Pull in block-5.7 fixes for 5.8. Mostly to resolve a conflict with
the blk-iocost changes, but we also need the base of the bdi
use-after-free as well as we build on top of it.

* block-5.7:
  nvme: fix possible hang when ns scanning fails during error recovery
  nvme-pci: fix "slimmer CQ head update"
  bdi: add a ->dev_name field to struct backing_dev_info
  bdi: use bdi_dev_name() to get device name
  bdi: move bdi_dev_name out of line
  vboxsf: don't use the source name in the bdi name
  iocost: protect iocg->abs_vdebt with iocg->waitq.lock
  block: remove the bd_openers checks in blk_drop_partitions
  nvme: prevent double free in nvme_alloc_ns() error handling
  null_blk: Cleanup zoned device initialization
  null_blk: Fix zoned command handling
  block: remove unused header
  blk-iocost: Fix error on iocost_ioc_vrate_adj
  bdev: Reduce time holding bd_mutex in sync in blkdev_close()
  buffer: remove useless comment and WB_REASON_FREE_MORE_MEM, reason.

Signed-off-by: Jens Axboe <[email protected]>

block/bfq-iosched.c

@@ -123,6 +123,7 @@
 #include <linux/ioprio.h>
 #include <linux/sbitmap.h>
 #include <linux/delay.h>
+#include <linux/backing-dev.h>
 
 #include "blk.h"
 #include "blk-mq.h"
@@ -4976,8 +4977,9 @@ bfq_set_next_ioprio_data(struct bfq_queue *bfqq, struct bfq_io_cq *bic)
 	ioprio_class = IOPRIO_PRIO_CLASS(bic->ioprio);
 	switch (ioprio_class) {
 	default:
-		dev_err(bfqq->bfqd->queue->backing_dev_info->dev,
-			"bfq: bad prio class %d\n", ioprio_class);
+		pr_err("bdi %s: bfq: bad prio class %d\n",
+				bdi_dev_name(bfqq->bfqd->queue->backing_dev_info),
+				ioprio_class);
 		/* fall through */
 	case IOPRIO_CLASS_NONE:
 		/*

block/blk-cgroup.c

@@ -496,7 +496,7 @@ const char *blkg_dev_name(struct blkcg_gq *blkg)
 {
 	/* some drivers (floppy) instantiate a queue w/o disk registered */
 	if (blkg->q->backing_dev_info->dev)
-		return dev_name(blkg->q->backing_dev_info->dev);
+		return bdi_dev_name(blkg->q->backing_dev_info);
 	return NULL;
 }
 

block/blk-iocost.c

@@ -467,7 +467,7 @@ struct ioc_gq {
 	 */
 	atomic64_t			vtime;
 	atomic64_t			done_vtime;
-	atomic64_t			abs_vdebt;
+	u64				abs_vdebt;
 	u64				last_vtime;
 
 	/*
@@ -1143,7 +1143,7 @@ static void iocg_kick_waitq(struct ioc_gq *iocg, struct ioc_now *now)
 	struct iocg_wake_ctx ctx = { .iocg = iocg };
 	u64 margin_ns = (u64)(ioc->period_us *
 			      WAITQ_TIMER_MARGIN_PCT / 100) * NSEC_PER_USEC;
-	u64 abs_vdebt, vdebt, vshortage, expires, oexpires;
+	u64 vdebt, vshortage, expires, oexpires;
 	s64 vbudget;
 	u32 hw_inuse;
 
@@ -1153,18 +1153,15 @@ static void iocg_kick_waitq(struct ioc_gq *iocg, struct ioc_now *now)
 	vbudget = now->vnow - atomic64_read(&iocg->vtime);
 
 	/* pay off debt */
-	abs_vdebt = atomic64_read(&iocg->abs_vdebt);
-	vdebt = abs_cost_to_cost(abs_vdebt, hw_inuse);
+	vdebt = abs_cost_to_cost(iocg->abs_vdebt, hw_inuse);
 	if (vdebt && vbudget > 0) {
 		u64 delta = min_t(u64, vbudget, vdebt);
 		u64 abs_delta = min(cost_to_abs_cost(delta, hw_inuse),
-				    abs_vdebt);
+				    iocg->abs_vdebt);
 
 		atomic64_add(delta, &iocg->vtime);
 		atomic64_add(delta, &iocg->done_vtime);
-		atomic64_sub(abs_delta, &iocg->abs_vdebt);
-		if (WARN_ON_ONCE(atomic64_read(&iocg->abs_vdebt) < 0))
-			atomic64_set(&iocg->abs_vdebt, 0);
+		iocg->abs_vdebt -= abs_delta;
 	}
 
 	/*
@@ -1220,12 +1217,18 @@ static bool iocg_kick_delay(struct ioc_gq *iocg, struct ioc_now *now)
 	u64 delta_ns, expires, oexpires;
 	u32 hw_inuse;
 
+	lockdep_assert_held(&iocg->waitq.lock);
+
 	/* debt-adjust vtime */
 	current_hweight(iocg, NULL, &hw_inuse);
-	vtime += abs_cost_to_cost(atomic64_read(&iocg->abs_vdebt), hw_inuse);
+	vtime += abs_cost_to_cost(iocg->abs_vdebt, hw_inuse);
 
-	/* clear or maintain depending on the overage */
-	if (time_before_eq64(vtime, now->vnow)) {
+	/*
+	 * Clear or maintain depending on the overage. Non-zero vdebt is what
+	 * guarantees that @iocg is online and future iocg_kick_delay() will
+	 * clear use_delay. Don't leave it on when there's no vdebt.
+	 */
+	if (!iocg->abs_vdebt || time_before_eq64(vtime, now->vnow)) {
 		blkcg_clear_delay(blkg);
 		return false;
 	}
@@ -1254,9 +1257,12 @@ static enum hrtimer_restart iocg_delay_timer_fn(struct hrtimer *timer)
 {
 	struct ioc_gq *iocg = container_of(timer, struct ioc_gq, delay_timer);
 	struct ioc_now now;
+	unsigned long flags;
 
+	spin_lock_irqsave(&iocg->waitq.lock, flags);
 	ioc_now(iocg->ioc, &now);
 	iocg_kick_delay(iocg, &now);
+	spin_unlock_irqrestore(&iocg->waitq.lock, flags);
 
 	return HRTIMER_NORESTART;
 }
@@ -1364,14 +1370,13 @@ static void ioc_timer_fn(struct timer_list *timer)
 	 * should have woken up in the last period and expire idle iocgs.
 	 */
 	list_for_each_entry_safe(iocg, tiocg, &ioc->active_iocgs, active_list) {
-		if (!waitqueue_active(&iocg->waitq) &&
-		    !atomic64_read(&iocg->abs_vdebt) && !iocg_is_idle(iocg))
+		if (!waitqueue_active(&iocg->waitq) && iocg->abs_vdebt &&
+		    !iocg_is_idle(iocg))
 			continue;
 
 		spin_lock(&iocg->waitq.lock);
 
-		if (waitqueue_active(&iocg->waitq) ||
-		    atomic64_read(&iocg->abs_vdebt)) {
+		if (waitqueue_active(&iocg->waitq) || iocg->abs_vdebt) {
 			/* might be oversleeping vtime / hweight changes, kick */
 			iocg_kick_waitq(iocg, &now);
 			iocg_kick_delay(iocg, &now);
@@ -1587,7 +1592,7 @@ static void ioc_timer_fn(struct timer_list *timer)
 				      vrate_min, vrate_max);
 		}
 
-		trace_iocost_ioc_vrate_adj(ioc, vrate, &missed_ppm, rq_wait_pct,
+		trace_iocost_ioc_vrate_adj(ioc, vrate, missed_ppm, rq_wait_pct,
 					   nr_lagging, nr_shortages,
 					   nr_surpluses);
 
@@ -1596,7 +1601,7 @@ static void ioc_timer_fn(struct timer_list *timer)
 			ioc->period_us * vrate * INUSE_MARGIN_PCT, 100);
 	} else if (ioc->busy_level != prev_busy_level || nr_lagging) {
 		trace_iocost_ioc_vrate_adj(ioc, atomic64_read(&ioc->vtime_rate),
-					   &missed_ppm, rq_wait_pct, nr_lagging,
+					   missed_ppm, rq_wait_pct, nr_lagging,
 					   nr_shortages, nr_surpluses);
 	}
 
@@ -1739,28 +1744,49 @@ static void ioc_rqos_throttle(struct rq_qos *rqos, struct bio *bio)
 	 * tests are racy but the races aren't systemic - we only miss once
 	 * in a while which is fine.
 	 */
-	if (!waitqueue_active(&iocg->waitq) &&
-	    !atomic64_read(&iocg->abs_vdebt) &&
+	if (!waitqueue_active(&iocg->waitq) && !iocg->abs_vdebt &&
 	    time_before_eq64(vtime + cost, now.vnow)) {
 		iocg_commit_bio(iocg, bio, cost);
 		return;
 	}
 
 	/*
-	 * We're over budget.  If @bio has to be issued regardless,
-	 * remember the abs_cost instead of advancing vtime.
-	 * iocg_kick_waitq() will pay off the debt before waking more IOs.
+	 * We activated above but w/o any synchronization. Deactivation is
+	 * synchronized with waitq.lock and we won't get deactivated as long
+	 * as we're waiting or has debt, so we're good if we're activated
+	 * here. In the unlikely case that we aren't, just issue the IO.
+	 */
+	spin_lock_irq(&iocg->waitq.lock);
+
+	if (unlikely(list_empty(&iocg->active_list))) {
+		spin_unlock_irq(&iocg->waitq.lock);
+		iocg_commit_bio(iocg, bio, cost);
+		return;
+	}
+
+	/*
+	 * We're over budget. If @bio has to be issued regardless, remember
+	 * the abs_cost instead of advancing vtime. iocg_kick_waitq() will pay
+	 * off the debt before waking more IOs.
+	 *
 	 * This way, the debt is continuously paid off each period with the
-	 * actual budget available to the cgroup.  If we just wound vtime,
-	 * we would incorrectly use the current hw_inuse for the entire
-	 * amount which, for example, can lead to the cgroup staying
-	 * blocked for a long time even with substantially raised hw_inuse.
+	 * actual budget available to the cgroup. If we just wound vtime, we
+	 * would incorrectly use the current hw_inuse for the entire amount
+	 * which, for example, can lead to the cgroup staying blocked for a
+	 * long time even with substantially raised hw_inuse.
+	 *
+	 * An iocg with vdebt should stay online so that the timer can keep
+	 * deducting its vdebt and [de]activate use_delay mechanism
+	 * accordingly. We don't want to race against the timer trying to
+	 * clear them and leave @iocg inactive w/ dangling use_delay heavily
+	 * penalizing the cgroup and its descendants.
 	 */
 	if (bio_issue_as_root_blkg(bio) || fatal_signal_pending(current)) {
-		atomic64_add(abs_cost, &iocg->abs_vdebt);
+		iocg->abs_vdebt += abs_cost;
 		if (iocg_kick_delay(iocg, &now))
 			blkcg_schedule_throttle(rqos->q,
 					(bio->bi_opf & REQ_SWAP) == REQ_SWAP);
+		spin_unlock_irq(&iocg->waitq.lock);
 		return;
 	}
 
@@ -1777,20 +1803,6 @@ static void ioc_rqos_throttle(struct rq_qos *rqos, struct bio *bio)
 	 * All waiters are on iocg->waitq and the wait states are
 	 * synchronized using waitq.lock.
 	 */
-	spin_lock_irq(&iocg->waitq.lock);
-
-	/*
-	 * We activated above but w/o any synchronization.  Deactivation is
-	 * synchronized with waitq.lock and we won't get deactivated as
-	 * long as we're waiting, so we're good if we're activated here.
-	 * In the unlikely case that we are deactivated, just issue the IO.
-	 */
-	if (unlikely(list_empty(&iocg->active_list))) {
-		spin_unlock_irq(&iocg->waitq.lock);
-		iocg_commit_bio(iocg, bio, cost);
-		return;
-	}
-
 	init_waitqueue_func_entry(&wait.wait, iocg_wake_fn);
 	wait.wait.private = current;
 	wait.bio = bio;
@@ -1822,6 +1834,7 @@ static void ioc_rqos_merge(struct rq_qos *rqos, struct request *rq,
 	struct ioc_now now;
 	u32 hw_inuse;
 	u64 abs_cost, cost;
+	unsigned long flags;
 
 	/* bypass if disabled or for root cgroup */
 	if (!ioc->enabled || !iocg->level)
@@ -1841,15 +1854,28 @@ static void ioc_rqos_merge(struct rq_qos *rqos, struct request *rq,
 		iocg->cursor = bio_end;
 
 	/*
-	 * Charge if there's enough vtime budget and the existing request
-	 * has cost assigned.  Otherwise, account it as debt.  See debt
-	 * handling in ioc_rqos_throttle() for details.
+	 * Charge if there's enough vtime budget and the existing request has
+	 * cost assigned.
 	 */
 	if (rq->bio && rq->bio->bi_iocost_cost &&
-	    time_before_eq64(atomic64_read(&iocg->vtime) + cost, now.vnow))
+	    time_before_eq64(atomic64_read(&iocg->vtime) + cost, now.vnow)) {
 		iocg_commit_bio(iocg, bio, cost);
-	else
-		atomic64_add(abs_cost, &iocg->abs_vdebt);
+		return;
+	}
+
+	/*
+	 * Otherwise, account it as debt if @iocg is online, which it should
+	 * be for the vast majority of cases. See debt handling in
+	 * ioc_rqos_throttle() for details.
+	 */
+	spin_lock_irqsave(&iocg->waitq.lock, flags);
+	if (likely(!list_empty(&iocg->active_list))) {
+		iocg->abs_vdebt += abs_cost;
+		iocg_kick_delay(iocg, &now);
+	} else {
+		iocg_commit_bio(iocg, bio, cost);
+	}
+	spin_unlock_irqrestore(&iocg->waitq.lock, flags);
 }
 
 static void ioc_rqos_done_bio(struct rq_qos *rqos, struct bio *bio)
@@ -2021,7 +2047,6 @@ static void ioc_pd_init(struct blkg_policy_data *pd)
 	iocg->ioc = ioc;
 	atomic64_set(&iocg->vtime, now.vnow);
 	atomic64_set(&iocg->done_vtime, now.vnow);
-	atomic64_set(&iocg->abs_vdebt, 0);
 	atomic64_set(&iocg->active_period, atomic64_read(&ioc->cur_period));
 	INIT_LIST_HEAD(&iocg->active_list);
 	iocg->hweight_active = HWEIGHT_WHOLE;

block/partitions/core.c

@@ -610,7 +610,7 @@ int blk_drop_partitions(struct block_device *bdev)
 
 	if (!disk_part_scan_enabled(bdev->bd_disk))
 		return 0;
-	if (bdev->bd_part_count || bdev->bd_openers > 1)
+	if (bdev->bd_part_count)
 		return -EBUSY;
 
 	sync_blockdev(bdev);

drivers/block/null_blk.h

@@ -85,26 +85,35 @@ struct nullb {
 	char disk_name[DISK_NAME_LEN];
 };
 
+blk_status_t null_process_cmd(struct nullb_cmd *cmd,
+			      enum req_opf op, sector_t sector,
+			      unsigned int nr_sectors);
+
 #ifdef CONFIG_BLK_DEV_ZONED
-int null_zone_init(struct nullb_device *dev);
-void null_zone_exit(struct nullb_device *dev);
+int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q);
+int null_register_zoned_dev(struct nullb *nullb);
+void null_free_zoned_dev(struct nullb_device *dev);
 int null_report_zones(struct gendisk *disk, sector_t sector,
 		      unsigned int nr_zones, report_zones_cb cb, void *data);
-blk_status_t null_handle_zoned(struct nullb_cmd *cmd,
-				enum req_opf op, sector_t sector,
-				sector_t nr_sectors);
+blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd,
+				    enum req_opf op, sector_t sector,
+				    sector_t nr_sectors);
 size_t null_zone_valid_read_len(struct nullb *nullb,
 				sector_t sector, unsigned int len);
 #else
-static inline int null_zone_init(struct nullb_device *dev)
+static inline int null_init_zoned_dev(struct nullb_device *dev,
+				      struct request_queue *q)
 {
 	pr_err("CONFIG_BLK_DEV_ZONED not enabled\n");
 	return -EINVAL;
 }
-static inline void null_zone_exit(struct nullb_device *dev) {}
-static inline blk_status_t null_handle_zoned(struct nullb_cmd *cmd,
-					     enum req_opf op, sector_t sector,
-					     sector_t nr_sectors)
+static inline int null_register_zoned_dev(struct nullb *nullb)
+{
+	return -ENODEV;
+}
+static inline void null_free_zoned_dev(struct nullb_device *dev) {}
+static inline blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd,
+			enum req_opf op, sector_t sector, sector_t nr_sectors)
 {
 	return BLK_STS_NOTSUPP;
 }