Merge tag 'block-6.5-2023-07-03' of git://git.kernel.dk/linux

Pull more block updates from Jens Axboe:
 "Mostly items that came in a bit late for the initial pull request,
  wanted to make sure they had the appropriate amount of linux-next soak
  before going upstream.

  Outside of stragglers, just generic fixes for either merge window
  items, or longer standing bugs"

* tag 'block-6.5-2023-07-03' of git://git.kernel.dk/linux: (25 commits)
  md/raid0: add discard support for the 'original' layout
  nvme: disable controller on reset state failure
  nvme: sync timeout work on failed reset
  nvme: ensure unquiesce on teardown
  cdrom/gdrom: Fix build error
  nvme: improved uring polling
  block: add request polling helper
  nvme-mpath: fix I/O failure with EAGAIN when failing over I/O
  nvme: host: fix command name spelling
  blk-sysfs: add a new attr_group for blk_mq
  blk-iocost: move wbt_enable/disable_default() out of spinlock
  blk-wbt: cleanup rwb_enabled() and wbt_disabled()
  blk-wbt: remove dead code to handle wbt enable/disable with io inflight
  blk-wbt: don't create wbt sysfs entry if CONFIG_BLK_WBT is disabled
  blk-mq: fix two misuses on RQF_USE_SCHED
  blk-throttle: Fix io statistics for cgroup v1
  bcache: Fix bcache device claiming
  bcache: Alloc holder object before async registration
  raid10: avoid spin_lock from fastpath from raid10_unplug()
  md: fix 'delete_mutex' deadlock
  ...

block/blk-cgroup.c

@@ -2086,6 +2086,9 @@ void blk_cgroup_bio_start(struct bio *bio)
 	struct blkg_iostat_set *bis;
 	unsigned long flags;
 
+	if (!cgroup_subsys_on_dfl(io_cgrp_subsys))
+		return;
+
 	/* Root-level stats are sourced from system-wide IO stats */
 	if (!cgroup_parent(blkcg->css.cgroup))
 		return;
@@ -2116,8 +2119,7 @@ void blk_cgroup_bio_start(struct bio *bio)
 	}
 
 	u64_stats_update_end_irqrestore(&bis->sync, flags);
-	if (cgroup_subsys_on_dfl(io_cgrp_subsys))
-		cgroup_rstat_updated(blkcg->css.cgroup, cpu);
+	cgroup_rstat_updated(blkcg->css.cgroup, cpu);
 	put_cpu();
 }
 

block/blk-iocost.c

@@ -3301,11 +3301,9 @@ static ssize_t ioc_qos_write(struct kernfs_open_file *of, char *input,
 		blk_stat_enable_accounting(disk->queue);
 		blk_queue_flag_set(QUEUE_FLAG_RQ_ALLOC_TIME, disk->queue);
 		ioc->enabled = true;
-		wbt_disable_default(disk);
 	} else {
 		blk_queue_flag_clear(QUEUE_FLAG_RQ_ALLOC_TIME, disk->queue);
 		ioc->enabled = false;
-		wbt_enable_default(disk);
 	}
 
 	if (user) {
@@ -3318,6 +3316,11 @@ static ssize_t ioc_qos_write(struct kernfs_open_file *of, char *input,
 	ioc_refresh_params(ioc, true);
 	spin_unlock_irq(&ioc->lock);
 
+	if (enable)
+		wbt_disable_default(disk);
+	else
+		wbt_enable_default(disk);
+
 	blk_mq_unquiesce_queue(disk->queue);
 	blk_mq_unfreeze_queue(disk->queue);
 

block/blk-mq.c

@@ -49,17 +49,8 @@ static void blk_mq_request_bypass_insert(struct request *rq,
 		blk_insert_t flags);
 static void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx,
 		struct list_head *list);
-
-static inline struct blk_mq_hw_ctx *blk_qc_to_hctx(struct request_queue *q,
-		blk_qc_t qc)
-{
-	return xa_load(&q->hctx_table, qc);
-}
-
-static inline blk_qc_t blk_rq_to_qc(struct request *rq)
-{
-	return rq->mq_hctx->queue_num;
-}
+static int blk_hctx_poll(struct request_queue *q, struct blk_mq_hw_ctx *hctx,
+			 struct io_comp_batch *iob, unsigned int flags);
 
 /*
  * Check if any of the ctx, dispatch list or elevator
@@ -1248,7 +1239,7 @@ void blk_mq_start_request(struct request *rq)
 		q->integrity.profile->prepare_fn(rq);
 #endif
 	if (rq->bio && rq->bio->bi_opf & REQ_POLLED)
-	        WRITE_ONCE(rq->bio->bi_cookie, blk_rq_to_qc(rq));
+	        WRITE_ONCE(rq->bio->bi_cookie, rq->mq_hctx->queue_num);
 }
 EXPORT_SYMBOL(blk_mq_start_request);
 
@@ -1280,7 +1271,11 @@ static void blk_add_rq_to_plug(struct blk_plug *plug, struct request *rq)
 
 	if (!plug->multiple_queues && last && last->q != rq->q)
 		plug->multiple_queues = true;
-	if (!plug->has_elevator && (rq->rq_flags & RQF_USE_SCHED))
+	/*
+	 * Any request allocated from sched tags can't be issued to
+	 * ->queue_rqs() directly
+	 */
+	if (!plug->has_elevator && (rq->rq_flags & RQF_SCHED_TAGS))
 		plug->has_elevator = true;
 	rq->rq_next = NULL;
 	rq_list_add(&plug->mq_list, rq);
@@ -1350,7 +1345,7 @@ EXPORT_SYMBOL_GPL(blk_rq_is_poll);
 static void blk_rq_poll_completion(struct request *rq, struct completion *wait)
 {
 	do {
-		blk_mq_poll(rq->q, blk_rq_to_qc(rq), NULL, 0);
+		blk_hctx_poll(rq->q, rq->mq_hctx, NULL, 0);
 		cond_resched();
 	} while (!completion_done(wait));
 }
@@ -4745,10 +4740,9 @@ void blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, int nr_hw_queues)
 }
 EXPORT_SYMBOL_GPL(blk_mq_update_nr_hw_queues);
 
-int blk_mq_poll(struct request_queue *q, blk_qc_t cookie, struct io_comp_batch *iob,
-		unsigned int flags)
+static int blk_hctx_poll(struct request_queue *q, struct blk_mq_hw_ctx *hctx,
+			 struct io_comp_batch *iob, unsigned int flags)
 {
-	struct blk_mq_hw_ctx *hctx = blk_qc_to_hctx(q, cookie);
 	long state = get_current_state();
 	int ret;
 
@@ -4773,6 +4767,32 @@ int blk_mq_poll(struct request_queue *q, blk_qc_t cookie, struct io_comp_batch *
 	return 0;
 }
 
+int blk_mq_poll(struct request_queue *q, blk_qc_t cookie,
+		struct io_comp_batch *iob, unsigned int flags)
+{
+	struct blk_mq_hw_ctx *hctx = xa_load(&q->hctx_table, cookie);
+
+	return blk_hctx_poll(q, hctx, iob, flags);
+}
+
+int blk_rq_poll(struct request *rq, struct io_comp_batch *iob,
+		unsigned int poll_flags)
+{
+	struct request_queue *q = rq->q;
+	int ret;
+
+	if (!blk_rq_is_poll(rq))
+		return 0;
+	if (!percpu_ref_tryget(&q->q_usage_counter))
+		return 0;
+
+	ret = blk_hctx_poll(q, rq->mq_hctx, iob, poll_flags);
+	blk_queue_exit(q);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(blk_rq_poll);
+
 unsigned int blk_mq_rq_cpu(struct request *rq)
 {
 	return rq->mq_ctx->cpu;

block/blk-sysfs.c

@@ -47,19 +47,6 @@ queue_var_store(unsigned long *var, const char *page, size_t count)
 	return count;
 }
 
-static ssize_t queue_var_store64(s64 *var, const char *page)
-{
-	int err;
-	s64 v;
-
-	err = kstrtos64(page, 10, &v);
-	if (err < 0)
-		return err;
-
-	*var = v;
-	return 0;
-}
-
 static ssize_t queue_requests_show(struct request_queue *q, char *page)
 {
 	return queue_var_show(q->nr_requests, page);
@@ -451,61 +438,6 @@ static ssize_t queue_io_timeout_store(struct request_queue *q, const char *page,
 	return count;
 }
 
-static ssize_t queue_wb_lat_show(struct request_queue *q, char *page)
-{
-	if (!wbt_rq_qos(q))
-		return -EINVAL;
-
-	if (wbt_disabled(q))
-		return sprintf(page, "0\n");
-
-	return sprintf(page, "%llu\n", div_u64(wbt_get_min_lat(q), 1000));
-}
-
-static ssize_t queue_wb_lat_store(struct request_queue *q, const char *page,
-				  size_t count)
-{
-	struct rq_qos *rqos;
-	ssize_t ret;
-	s64 val;
-
-	ret = queue_var_store64(&val, page);
-	if (ret < 0)
-		return ret;
-	if (val < -1)
-		return -EINVAL;
-
-	rqos = wbt_rq_qos(q);
-	if (!rqos) {
-		ret = wbt_init(q->disk);
-		if (ret)
-			return ret;
-	}
-
-	if (val == -1)
-		val = wbt_default_latency_nsec(q);
-	else if (val >= 0)
-		val *= 1000ULL;
-
-	if (wbt_get_min_lat(q) == val)
-		return count;
-
-	/*
-	 * Ensure that the queue is idled, in case the latency update
-	 * ends up either enabling or disabling wbt completely. We can't
-	 * have IO inflight if that happens.
-	 */
-	blk_mq_freeze_queue(q);
-	blk_mq_quiesce_queue(q);
-
-	wbt_set_min_lat(q, val);
-
-	blk_mq_unquiesce_queue(q);
-	blk_mq_unfreeze_queue(q);
-
-	return count;
-}
-
 static ssize_t queue_wc_show(struct request_queue *q, char *page)
 {
 	if (test_bit(QUEUE_FLAG_WC, &q->queue_flags))
@@ -598,7 +530,6 @@ QUEUE_RW_ENTRY(queue_wc, "write_cache");
 QUEUE_RO_ENTRY(queue_fua, "fua");
 QUEUE_RO_ENTRY(queue_dax, "dax");
 QUEUE_RW_ENTRY(queue_io_timeout, "io_timeout");
-QUEUE_RW_ENTRY(queue_wb_lat, "wbt_lat_usec");
 QUEUE_RO_ENTRY(queue_virt_boundary_mask, "virt_boundary_mask");
 QUEUE_RO_ENTRY(queue_dma_alignment, "dma_alignment");
 
@@ -617,8 +548,79 @@ QUEUE_RW_ENTRY(queue_iostats, "iostats");
 QUEUE_RW_ENTRY(queue_random, "add_random");
 QUEUE_RW_ENTRY(queue_stable_writes, "stable_writes");
 
+#ifdef CONFIG_BLK_WBT
+static ssize_t queue_var_store64(s64 *var, const char *page)
+{
+	int err;
+	s64 v;
+
+	err = kstrtos64(page, 10, &v);
+	if (err < 0)
+		return err;
+
+	*var = v;
+	return 0;
+}
+
+static ssize_t queue_wb_lat_show(struct request_queue *q, char *page)
+{
+	if (!wbt_rq_qos(q))
+		return -EINVAL;
+
+	if (wbt_disabled(q))
+		return sprintf(page, "0\n");
+
+	return sprintf(page, "%llu\n", div_u64(wbt_get_min_lat(q), 1000));
+}
+
+static ssize_t queue_wb_lat_store(struct request_queue *q, const char *page,
+				  size_t count)
+{
+	struct rq_qos *rqos;
+	ssize_t ret;
+	s64 val;
+
+	ret = queue_var_store64(&val, page);
+	if (ret < 0)
+		return ret;
+	if (val < -1)
+		return -EINVAL;
+
+	rqos = wbt_rq_qos(q);
+	if (!rqos) {
+		ret = wbt_init(q->disk);
+		if (ret)
+			return ret;
+	}
+
+	if (val == -1)
+		val = wbt_default_latency_nsec(q);
+	else if (val >= 0)
+		val *= 1000ULL;
+
+	if (wbt_get_min_lat(q) == val)
+		return count;
+
+	/*
+	 * Ensure that the queue is idled, in case the latency update
+	 * ends up either enabling or disabling wbt completely. We can't
+	 * have IO inflight if that happens.
+	 */
+	blk_mq_freeze_queue(q);
+	blk_mq_quiesce_queue(q);
+
+	wbt_set_min_lat(q, val);
+
+	blk_mq_unquiesce_queue(q);
+	blk_mq_unfreeze_queue(q);
+
+	return count;
+}
+
+QUEUE_RW_ENTRY(queue_wb_lat, "wbt_lat_usec");
+#endif
+
 static struct attribute *queue_attrs[] = {
-	&queue_requests_entry.attr,
 	&queue_ra_entry.attr,
 	&queue_max_hw_sectors_entry.attr,
 	&queue_max_sectors_entry.attr,
@@ -626,7 +628,6 @@ static struct attribute *queue_attrs[] = {
 	&queue_max_discard_segments_entry.attr,
 	&queue_max_integrity_segments_entry.attr,
 	&queue_max_segment_size_entry.attr,
-	&elv_iosched_entry.attr,
 	&queue_hw_sector_size_entry.attr,
 	&queue_logical_block_size_entry.attr,
 	&queue_physical_block_size_entry.attr,
@@ -647,7 +648,6 @@ static struct attribute *queue_attrs[] = {
 	&queue_max_open_zones_entry.attr,
 	&queue_max_active_zones_entry.attr,
 	&queue_nomerges_entry.attr,
-	&queue_rq_affinity_entry.attr,
 	&queue_iostats_entry.attr,
 	&queue_stable_writes_entry.attr,
 	&queue_random_entry.attr,
@@ -655,9 +655,7 @@ static struct attribute *queue_attrs[] = {
 	&queue_wc_entry.attr,
 	&queue_fua_entry.attr,
 	&queue_dax_entry.attr,
-	&queue_wb_lat_entry.attr,
 	&queue_poll_delay_entry.attr,
-	&queue_io_timeout_entry.attr,
 #ifdef CONFIG_BLK_DEV_THROTTLING_LOW
 	&blk_throtl_sample_time_entry.attr,
 #endif
@@ -666,16 +664,23 @@ static struct attribute *queue_attrs[] = {
 	NULL,
 };
 
+static struct attribute *blk_mq_queue_attrs[] = {
+	&queue_requests_entry.attr,
+	&elv_iosched_entry.attr,
+	&queue_rq_affinity_entry.attr,
+	&queue_io_timeout_entry.attr,
+#ifdef CONFIG_BLK_WBT
+	&queue_wb_lat_entry.attr,
+#endif
+	NULL,
+};
+
 static umode_t queue_attr_visible(struct kobject *kobj, struct attribute *attr,
 				int n)
 {
 	struct gendisk *disk = container_of(kobj, struct gendisk, queue_kobj);
 	struct request_queue *q = disk->queue;
 
-	if (attr == &queue_io_timeout_entry.attr &&
-		(!q->mq_ops || !q->mq_ops->timeout))
-			return 0;
-
 	if ((attr == &queue_max_open_zones_entry.attr ||
 	     attr == &queue_max_active_zones_entry.attr) &&
 	    !blk_queue_is_zoned(q))
@@ -684,11 +689,30 @@ static umode_t queue_attr_visible(struct kobject *kobj, struct attribute *attr,
 	return attr->mode;
 }
 
+static umode_t blk_mq_queue_attr_visible(struct kobject *kobj,
+					 struct attribute *attr, int n)
+{
+	struct gendisk *disk = container_of(kobj, struct gendisk, queue_kobj);
+	struct request_queue *q = disk->queue;
+
+	if (!queue_is_mq(q))
+		return 0;
+
+	if (attr == &queue_io_timeout_entry.attr && !q->mq_ops->timeout)
+		return 0;
+
+	return attr->mode;
+}
+
 static struct attribute_group queue_attr_group = {
 	.attrs = queue_attrs,
 	.is_visible = queue_attr_visible,
 };
 
+static struct attribute_group blk_mq_queue_attr_group = {
+	.attrs = blk_mq_queue_attrs,
+	.is_visible = blk_mq_queue_attr_visible,
+};
 
 #define to_queue(atr) container_of((atr), struct queue_sysfs_entry, attr)
 
@@ -733,6 +757,7 @@ static const struct sysfs_ops queue_sysfs_ops = {
 
 static const struct attribute_group *blk_queue_attr_groups[] = {
 	&queue_attr_group,
+	&blk_mq_queue_attr_group,
 	NULL
 };
 

block/blk-throttle.c

@@ -2178,12 +2178,6 @@ bool __blk_throtl_bio(struct bio *bio)
 
 	rcu_read_lock();
 
-	if (!cgroup_subsys_on_dfl(io_cgrp_subsys)) {
-		blkg_rwstat_add(&tg->stat_bytes, bio->bi_opf,
-				bio->bi_iter.bi_size);
-		blkg_rwstat_add(&tg->stat_ios, bio->bi_opf, 1);
-	}
-
 	spin_lock_irq(&q->queue_lock);
 
 	throtl_update_latency_buckets(td);

block/blk-throttle.h

@@ -185,6 +185,15 @@ static inline bool blk_should_throtl(struct bio *bio)
 	struct throtl_grp *tg = blkg_to_tg(bio->bi_blkg);
 	int rw = bio_data_dir(bio);
 
+	if (!cgroup_subsys_on_dfl(io_cgrp_subsys)) {
+		if (!bio_flagged(bio, BIO_CGROUP_ACCT)) {
+			bio_set_flag(bio, BIO_CGROUP_ACCT);
+			blkg_rwstat_add(&tg->stat_bytes, bio->bi_opf,
+					bio->bi_iter.bi_size);
+		}
+		blkg_rwstat_add(&tg->stat_ios, bio->bi_opf, 1);
+	}
+
 	/* iops limit is always counted */
 	if (tg->has_rules_iops[rw])
 		return true;

block/blk-wbt.c

@@ -146,7 +146,7 @@ enum {
 static inline bool rwb_enabled(struct rq_wb *rwb)
 {
 	return rwb && rwb->enable_state != WBT_STATE_OFF_DEFAULT &&
-		      rwb->wb_normal != 0;
+		      rwb->enable_state != WBT_STATE_OFF_MANUAL;
 }
 
 static void wb_timestamp(struct rq_wb *rwb, unsigned long *var)
@@ -200,15 +200,6 @@ static void wbt_rqw_done(struct rq_wb *rwb, struct rq_wait *rqw,
 
 	inflight = atomic_dec_return(&rqw->inflight);
 
-	/*
-	 * wbt got disabled with IO in flight. Wake up any potential
-	 * waiters, we don't have to do more than that.
-	 */
-	if (unlikely(!rwb_enabled(rwb))) {
-		rwb_wake_all(rwb);
-		return;
-	}
-
 	/*
 	 * For discards, our limit is always the background. For writes, if
 	 * the device does write back caching, drop further down before we
@@ -503,8 +494,7 @@ bool wbt_disabled(struct request_queue *q)
 {
 	struct rq_qos *rqos = wbt_rq_qos(q);
 
-	return !rqos || RQWB(rqos)->enable_state == WBT_STATE_OFF_DEFAULT ||
-	       RQWB(rqos)->enable_state == WBT_STATE_OFF_MANUAL;
+	return !rqos || !rwb_enabled(RQWB(rqos));
 }
 
 u64 wbt_get_min_lat(struct request_queue *q)
@@ -545,13 +535,6 @@ static inline unsigned int get_limit(struct rq_wb *rwb, blk_opf_t opf)
 {
 	unsigned int limit;
 
-	/*
-	 * If we got disabled, just return UINT_MAX. This ensures that
-	 * we'll properly inc a new IO, and dec+wakeup at the end.
-	 */
-	if (!rwb_enabled(rwb))
-		return UINT_MAX;
-
 	if ((opf & REQ_OP_MASK) == REQ_OP_DISCARD)
 		return rwb->wb_background;
 

block/blk-wbt.h

@@ -18,10 +18,6 @@ u64 wbt_default_latency_nsec(struct request_queue *);
 
 #else
 
-static inline int wbt_init(struct gendisk *disk)
-{
-	return -EINVAL;
-}
 static inline void wbt_disable_default(struct gendisk *disk)
 {
 }
@@ -31,21 +27,6 @@ static inline void wbt_enable_default(struct gendisk *disk)
 static inline void wbt_set_write_cache(struct request_queue *q, bool wc)
 {
 }
-static inline u64 wbt_get_min_lat(struct request_queue *q)
-{
-	return 0;
-}
-static inline void wbt_set_min_lat(struct request_queue *q, u64 val)
-{
-}
-static inline u64 wbt_default_latency_nsec(struct request_queue *q)
-{
-	return 0;
-}
-static inline bool wbt_disabled(struct request_queue *q)
-{
-	return true;
-}
 
 #endif /* CONFIG_BLK_WBT */