Merge tag 'block-5.6-2020-03-13' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A few fixes that should go into this release. This contains:

   - Fix for a corruption issue with the s390 dasd driver (Stefan)

   - Fixup/improvement for the flush insertion change that we had in
     this series (Ming)

   - Fix for the partition suppor for host aware zoned devices
     (Shin'ichiro)

   - Fix incorrect blk-iocost comparison (Tejun)

  The diffstat looks large, but that's a) mostly dasd, and b) the flush
  fix from Ming adds a big comment"

* tag 'block-5.6-2020-03-13' of git://git.kernel.dk/linux-block:
  block: Fix partition support for host aware zoned block devices
  blk-mq: insert flush request to the front of dispatch queue
  s390/dasd: fix data corruption for thin provisioned devices
  blk-iocost: fix incorrect vtime comparison in iocg_is_idle()

7 files changed, 253 insertions, 25 deletions

diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 27ca68621137..9a599cc28c29 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -1318,7 +1318,7 @@ static bool iocg_is_idle(struct ioc_gq *iocg)
 		return false;
 
 	/* is something in flight? */
-	if (atomic64_read(&iocg->done_vtime) < atomic64_read(&iocg->vtime))
+	if (atomic64_read(&iocg->done_vtime) != atomic64_read(&iocg->vtime))
 		return false;
 
 	return true;
diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index 856356b1619e..74cedea56034 100644
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -398,6 +398,28 @@ void blk_mq_sched_insert_request(struct request *rq, bool at_head,
 	WARN_ON(e && (rq->tag != -1));
 
 	if (blk_mq_sched_bypass_insert(hctx, !!e, rq)) {
+		/*
+		 * Firstly normal IO request is inserted to scheduler queue or
+		 * sw queue, meantime we add flush request to dispatch queue(
+		 * hctx->dispatch) directly and there is at most one in-flight
+		 * flush request for each hw queue, so it doesn't matter to add
+		 * flush request to tail or front of the dispatch queue.
+		 *
+		 * Secondly in case of NCQ, flush request belongs to non-NCQ
+		 * command, and queueing it will fail when there is any
+		 * in-flight normal IO request(NCQ command). When adding flush
+		 * rq to the front of hctx->dispatch, it is easier to introduce
+		 * extra time to flush rq's latency because of S_SCHED_RESTART
+		 * compared with adding to the tail of dispatch queue, then
+		 * chance of flush merge is increased, and less flush requests
+		 * will be issued to controller. It is observed that ~10% time
+		 * is saved in blktests block/004 on disk attached to AHCI/NCQ
+		 * drive when adding flush rq to the front of hctx->dispatch.
+		 *
+		 * Simply queue flush rq to the front of hctx->dispatch so that
+		 * intensive flush workloads can benefit in case of NCQ HW.
+		 */
+		at_head = (rq->rq_flags & RQF_FLUSH_SEQ) ? true : at_head;
 		blk_mq_request_bypass_insert(rq, at_head, false);
 		goto run;
 	}
diff --git a/block/genhd.c b/block/genhd.c
index ff6268970ddc..9c2e13ce0d19 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -301,6 +301,42 @@ struct hd_struct *disk_map_sector_rcu(struct gendisk *disk, sector_t sector)
 }
 EXPORT_SYMBOL_GPL(disk_map_sector_rcu);
 
+/**
+ * disk_has_partitions
+ * @disk: gendisk of interest
+ *
+ * Walk through the partition table and check if valid partition exists.
+ *
+ * CONTEXT:
+ * Don't care.
+ *
+ * RETURNS:
+ * True if the gendisk has at least one valid non-zero size partition.
+ * Otherwise false.
+ */
+bool disk_has_partitions(struct gendisk *disk)
+{
+	struct disk_part_tbl *ptbl;
+	int i;
+	bool ret = false;
+
+	rcu_read_lock();
+	ptbl = rcu_dereference(disk->part_tbl);
+
+	/* Iterate partitions skipping the whole device at index 0 */
+	for (i = 1; i < ptbl->len; i++) {
+		if (rcu_dereference(ptbl->part[i])) {
+			ret = true;
+			break;
+		}
+	}
+
+	rcu_read_unlock();
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(disk_has_partitions);
+
 /*
  * Can be deleted altogether. Later.
  *
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 6cca72782af6..cf87eb27879f 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -178,6 +178,8 @@ struct dasd_block *dasd_alloc_block(void)
 		     (unsigned long) block);
 	INIT_LIST_HEAD(&block->ccw_queue);
 	spin_lock_init(&block->queue_lock);
+	INIT_LIST_HEAD(&block->format_list);
+	spin_lock_init(&block->format_lock);
 	timer_setup(&block->timer, dasd_block_timeout, 0);
 	spin_lock_init(&block->profile.lock);
 
@@ -1779,20 +1781,26 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
 
 	if (dasd_ese_needs_format(cqr->block, irb)) {
 		if (rq_data_dir((struct request *)cqr->callback_data) == READ) {
-			device->discipline->ese_read(cqr);
+			device->discipline->ese_read(cqr, irb);
 			cqr->status = DASD_CQR_SUCCESS;
 			cqr->stopclk = now;
 			dasd_device_clear_timer(device);
 			dasd_schedule_device_bh(device);
 			return;
 		}
-		fcqr = device->discipline->ese_format(device, cqr);
+		fcqr = device->discipline->ese_format(device, cqr, irb);
 		if (IS_ERR(fcqr)) {
+			if (PTR_ERR(fcqr) == -EINVAL) {
+				cqr->status = DASD_CQR_ERROR;
+				return;
+			}
 			/*
 			 * If we can't format now, let the request go
 			 * one extra round. Maybe we can format later.
 			 */
 			cqr->status = DASD_CQR_QUEUED;
+			dasd_schedule_device_bh(device);
+			return;
 		} else {
 			fcqr->status = DASD_CQR_QUEUED;
 			cqr->status = DASD_CQR_QUEUED;
@@ -2748,11 +2756,13 @@ static void __dasd_cleanup_cqr(struct dasd_ccw_req *cqr)
 {
 	struct request *req;
 	blk_status_t error = BLK_STS_OK;
+	unsigned int proc_bytes;
 	int status;
 
 	req = (struct request *) cqr->callback_data;
 	dasd_profile_end(cqr->block, cqr, req);
 
+	proc_bytes = cqr->proc_bytes;
 	status = cqr->block->base->discipline->free_cp(cqr, req);
 	if (status < 0)
 		error = errno_to_blk_status(status);
@@ -2783,7 +2793,18 @@ static void __dasd_cleanup_cqr(struct dasd_ccw_req *cqr)
 		blk_mq_end_request(req, error);
 		blk_mq_run_hw_queues(req->q, true);
 	} else {
-		blk_mq_complete_request(req);
+		/*
+		 * Partial completed requests can happen with ESE devices.
+		 * During read we might have gotten a NRF error and have to
+		 * complete a request partially.
+		 */
+		if (proc_bytes) {
+			blk_update_request(req, BLK_STS_OK,
+					   blk_rq_bytes(req) - proc_bytes);
+			blk_mq_requeue_request(req, true);
+		} else {
+			blk_mq_complete_request(req);
+		}
 	}
 }
 
diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c
index a28b9ff82378..ad44d22e8859 100644
--- a/drivers/s390/block/dasd_eckd.c
+++ b/drivers/s390/block/dasd_eckd.c
@@ -207,6 +207,45 @@ static void set_ch_t(struct ch_t *geo, __u32 cyl, __u8 head)
 	geo->head |= head;
 }
 
+/*
+ * calculate failing track from sense data depending if
+ * it is an EAV device or not
+ */
+static int dasd_eckd_track_from_irb(struct irb *irb, struct dasd_device *device,
+				    sector_t *track)
+{
+	struct dasd_eckd_private *private = device->private;
+	u8 *sense = NULL;
+	u32 cyl;
+	u8 head;
+
+	sense = dasd_get_sense(irb);
+	if (!sense) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "ESE error no sense data\n");
+		return -EINVAL;
+	}
+	if (!(sense[27] & DASD_SENSE_BIT_2)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "ESE error no valid track data\n");
+		return -EINVAL;
+	}
+
+	if (sense[27] & DASD_SENSE_BIT_3) {
+		/* enhanced addressing */
+		cyl = sense[30] << 20;
+		cyl |= (sense[31] & 0xF0) << 12;
+		cyl |= sense[28] << 8;
+		cyl |= sense[29];
+	} else {
+		cyl = sense[29] << 8;
+		cyl |= sense[30];
+	}
+	head = sense[31] & 0x0F;
+	*track = cyl * private->rdc_data.trk_per_cyl + head;
+	return 0;
+}
+
 static int set_timestamp(struct ccw1 *ccw, struct DE_eckd_data *data,
 		     struct dasd_device *device)
 {
@@ -2986,6 +3025,37 @@ static int dasd_eckd_format_device(struct dasd_device *base,
 					     0, NULL);
 }
 
+static bool test_and_set_format_track(struct dasd_format_entry *to_format,
+				      struct dasd_block *block)
+{
+	struct dasd_format_entry *format;
+	unsigned long flags;
+	bool rc = false;
+
+	spin_lock_irqsave(&block->format_lock, flags);
+	list_for_each_entry(format, &block->format_list, list) {
+		if (format->track == to_format->track) {
+			rc = true;
+			goto out;
+		}
+	}
+	list_add_tail(&to_format->list, &block->format_list);
+
+out:
+	spin_unlock_irqrestore(&block->format_lock, flags);
+	return rc;
+}
+
+static void clear_format_track(struct dasd_format_entry *format,
+			      struct dasd_block *block)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&block->format_lock, flags);
+	list_del_init(&format->list);
+	spin_unlock_irqrestore(&block->format_lock, flags);
+}
+
 /*
  * Callback function to free ESE format requests.
  */
@@ -2993,15 +3063,19 @@ static void dasd_eckd_ese_format_cb(struct dasd_ccw_req *cqr, void *data)
 {
 	struct dasd_device *device = cqr->startdev;
 	struct dasd_eckd_private *private = device->private;
+	struct dasd_format_entry *format = data;
 
+	clear_format_track(format, cqr->basedev->block);
 	private->count--;
 	dasd_ffree_request(cqr, device);
 }
 
 static struct dasd_ccw_req *
-dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr)
+dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr,
+		     struct irb *irb)
 {
 	struct dasd_eckd_private *private;
+	struct dasd_format_entry *format;
 	struct format_data_t fdata;
 	unsigned int recs_per_trk;
 	struct dasd_ccw_req *fcqr;
@@ -3011,23 +3085,39 @@ dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr)
 	struct request *req;
 	sector_t first_trk;
 	sector_t last_trk;
+	sector_t curr_trk;
 	int rc;
 
 	req = cqr->callback_data;
-	base = cqr->block->base;
+	block = cqr->block;
+	base = block->base;
 	private = base->private;
-	block = base->block;
 	blksize = block->bp_block;
 	recs_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+	format = &startdev->format_entry;
 
 	first_trk = blk_rq_pos(req) >> block->s2b_shift;
 	sector_div(first_trk, recs_per_trk);
 	last_trk =
 		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
 	sector_div(last_trk, recs_per_trk);
+	rc = dasd_eckd_track_from_irb(irb, base, &curr_trk);
+	if (rc)
+		return ERR_PTR(rc);
 
-	fdata.start_unit = first_trk;
-	fdata.stop_unit = last_trk;
+	if (curr_trk < first_trk || curr_trk > last_trk) {
+		DBF_DEV_EVENT(DBF_WARNING, startdev,
+			      "ESE error track %llu not within range %llu - %llu\n",
+			      curr_trk, first_trk, last_trk);
+		return ERR_PTR(-EINVAL);
+	}
+	format->track = curr_trk;
+	/* test if track is already in formatting by another thread */
+	if (test_and_set_format_track(format, block))
+		return ERR_PTR(-EEXIST);
+
+	fdata.start_unit = curr_trk;
+	fdata.stop_unit = curr_trk;
 	fdata.blksize = blksize;
 	fdata.intensity = private->uses_cdl ? DASD_FMT_INT_COMPAT : 0;
 
@@ -3044,6 +3134,7 @@ dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr)
 		return fcqr;
 
 	fcqr->callback = dasd_eckd_ese_format_cb;
+	fcqr->callback_data = (void *) format;
 
 	return fcqr;
 }
@@ -3051,29 +3142,87 @@ dasd_eckd_ese_format(struct dasd_device *startdev, struct dasd_ccw_req *cqr)
 /*
  * When data is read from an unformatted area of an ESE volume, this function
  * returns zeroed data and thereby mimics a read of zero data.
+ *
+ * The first unformatted track is the one that got the NRF error, the address is
+ * encoded in the sense data.
+ *
+ * All tracks before have returned valid data and should not be touched.
+ * All tracks after the unformatted track might be formatted or not. This is
+ * currently not known, remember the processed data and return the remainder of
+ * the request to the blocklayer in __dasd_cleanup_cqr().
  */
-static void dasd_eckd_ese_read(struct dasd_ccw_req *cqr)
+static int dasd_eckd_ese_read(struct dasd_ccw_req *cqr, struct irb *irb)
 {
+	struct dasd_eckd_private *private;
+	sector_t first_trk, last_trk;
+	sector_t first_blk, last_blk;
 	unsigned int blksize, off;
+	unsigned int recs_per_trk;
 	struct dasd_device *base;
 	struct req_iterator iter;
+	struct dasd_block *block;
+	unsigned int skip_block;
+	unsigned int blk_count;
 	struct request *req;
 	struct bio_vec bv;
+	sector_t curr_trk;
+	sector_t end_blk;
 	char *dst;
+	int rc;
 
 	req = (struct request *) cqr->callback_data;
 	base = cqr->block->base;
 	blksize = base->block->bp_block;
+	block =  cqr->block;
+	private = base->private;
+	skip_block = 0;
+	blk_count = 0;
+
+	recs_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+	first_trk = first_blk = blk_rq_pos(req) >> block->s2b_shift;
+	sector_div(first_trk, recs_per_trk);
+	last_trk = last_blk =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	sector_div(last_trk, recs_per_trk);
+	rc = dasd_eckd_track_from_irb(irb, base, &curr_trk);
+	if (rc)
+		return rc;
+
+	/* sanity check if the current track from sense data is valid */
+	if (curr_trk < first_trk || curr_trk > last_trk) {
+		DBF_DEV_EVENT(DBF_WARNING, base,
+			      "ESE error track %llu not within range %llu - %llu\n",
+			      curr_trk, first_trk, last_trk);
+		return -EINVAL;
+	}
+
+	/*
+	 * if not the first track got the NRF error we have to skip over valid
+	 * blocks
+	 */
+	if (curr_trk != first_trk)
+		skip_block = curr_trk * recs_per_trk - first_blk;
+
+	/* we have no information beyond the current track */
+	end_blk = (curr_trk + 1) * recs_per_trk;
 
 	rq_for_each_segment(bv, req, iter) {
 		dst = page_address(bv.bv_page) + bv.bv_offset;
 		for (off = 0; off < bv.bv_len; off += blksize) {
-			if (dst && rq_data_dir(req) == READ) {
+			if (first_blk + blk_count >= end_blk) {
+				cqr->proc_bytes = blk_count * blksize;
+				return 0;
+			}
+			if (dst && !skip_block) {
 				dst += off;
 				memset(dst, 0, blksize);
+			} else {
+				skip_block--;
 			}
+			blk_count++;
 		}
 	}
+	return 0;
 }
 
 /*
diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h
index 91c9f9586e0f..fa552f9f1666 100644
--- a/drivers/s390/block/dasd_int.h
+++ b/drivers/s390/block/dasd_int.h
@@ -187,6 +187,7 @@ struct dasd_ccw_req {
 
 	void (*callback)(struct dasd_ccw_req *, void *data);
 	void *callback_data;
+	unsigned int proc_bytes;	/* bytes for partial completion */
 };
 
 /*
@@ -387,8 +388,9 @@ struct dasd_discipline {
 	int (*ext_pool_warn_thrshld)(struct dasd_device *);
 	int (*ext_pool_oos)(struct dasd_device *);
 	int (*ext_pool_exhaust)(struct dasd_device *, struct dasd_ccw_req *);
-	struct dasd_ccw_req *(*ese_format)(struct dasd_device *, struct dasd_ccw_req *);
-	void (*ese_read)(struct dasd_ccw_req *);
+	struct dasd_ccw_req *(*ese_format)(struct dasd_device *,
+					   struct dasd_ccw_req *, struct irb *);
+	int (*ese_read)(struct dasd_ccw_req *, struct irb *);
 };
 
 extern struct dasd_discipline *dasd_diag_discipline_pointer;
@@ -474,6 +476,11 @@ struct dasd_profile {
 	spinlock_t lock;
 };
 
+struct dasd_format_entry {
+	struct list_head list;
+	sector_t track;
+};
+
 struct dasd_device {
 	/* Block device stuff. */
 	struct dasd_block *block;
@@ -539,6 +546,7 @@ struct dasd_device {
 	struct dentry *debugfs_dentry;
 	struct dentry *hosts_dentry;
 	struct dasd_profile profile;
+	struct dasd_format_entry format_entry;
 };
 
 struct dasd_block {
@@ -564,6 +572,9 @@ struct dasd_block {
 
 	struct dentry *debugfs_dentry;
 	struct dasd_profile profile;
+
+	struct list_head format_list;
+	spinlock_t format_lock;
 };
 
 struct dasd_attention_data {
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 6fbe58538ad6..07dc91835b98 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -245,18 +245,6 @@ static inline bool disk_part_scan_enabled(struct gendisk *disk)
 		!(disk->flags & GENHD_FL_NO_PART_SCAN);
 }
 
-static inline bool disk_has_partitions(struct gendisk *disk)
-{
-	bool ret = false;
-
-	rcu_read_lock();
-	if (rcu_dereference(disk->part_tbl)->len > 1)
-		ret = true;
-	rcu_read_unlock();
-
-	return ret;
-}
-
 static inline dev_t disk_devt(struct gendisk *disk)
 {
 	return MKDEV(disk->major, disk->first_minor);
@@ -298,6 +286,7 @@ extern void disk_part_iter_exit(struct disk_part_iter *piter);
 
 extern struct hd_struct *disk_map_sector_rcu(struct gendisk *disk,
 					     sector_t sector);
+bool disk_has_partitions(struct gendisk *disk);
 
 /*
  * Macros to operate on percpu disk statistics: