Merge branch 'cached-zones' into for-6.19/block

This patch series implements a cached report zones using information
from the block layer zone write plugs and a new zone condition tracking.
This avoids having to execute slow report zones commands on the device
when for instance mounting file systems, which can significantly speed
things up, especially in setups with multiple SMR HDDs (e.g. a BTRFS
RAID volume).

The first patch improves handling of zone management commands. Patch 2
fixes zone resource updates and the following 3 patches cleanup the zone
code in preparation for introducing cached zone report support.
From patch 6 to 13, cached report zones is implemented and made
available to users with a new ioctl() command.

Finally, patches 14 and 15 introduce the use of cached report zones in
the mount operation of XFS and BTRFS.

Link: https://lore.kernel.org/linux-block/20251104212249.1075412-1-dlemoal@kernel.org/
Signed-off-by: Jens Axboe <axboe@kernel.dk>

* cached-zones:
  xfs: use blkdev_report_zones_cached()
  btrfs: use blkdev_report_zones_cached()
  block: add zone write plug condition to debugfs zone_wplugs
  block: improve zone_wplugs debugfs attribute output
  block: introduce BLKREPORTZONESV2 ioctl
  block: introduce blkdev_report_zones_cached()
  block: introduce blkdev_get_zone_info()
  block: refactor blkdev_report_zones() code
  block: track zone conditions
  block: use zone condition to determine conventional zones
  block: reorganize struct blk_zone_wplug
  block: introduce disk_report_zone()
  block: cleanup blkdev_report_zones()
  block: freeze queue when updating zone resources
  block: handle zone management operations completions

block/blk.h

@@ -489,9 +489,23 @@ static inline bool blk_req_bio_is_zone_append(struct request *rq,
 void blk_zone_write_plug_bio_merged(struct bio *bio);
 void blk_zone_write_plug_init_request(struct request *rq);
 void blk_zone_append_update_request_bio(struct request *rq, struct bio *bio);
+void blk_zone_mgmt_bio_endio(struct bio *bio);
 void blk_zone_write_plug_bio_endio(struct bio *bio);
 static inline void blk_zone_bio_endio(struct bio *bio)
 {
+	/*
+	 * Zone management BIOs may impact zone write plugs (e.g. a zone reset
+	 * changes a zone write plug zone write pointer offset), but these
+	 * operation do not go through zone write plugging as they may operate
+	 * on zones that do not have a zone write
+	 * plug. blk_zone_mgmt_bio_endio() handles the potential changes to zone
+	 * write plugs that are present.
+	 */
+	if (op_is_zone_mgmt(bio_op(bio))) {
+		blk_zone_mgmt_bio_endio(bio);
+		return;
+	}
+
 	/*
 	 * For write BIOs to zoned devices, signal the completion of the BIO so
 	 * that the next write BIO can be submitted by zone write plugging.

block/ioctl.c

@@ -581,6 +581,7 @@ static int blkdev_common_ioctl(struct block_device *bdev, blk_mode_t mode,
 	case BLKGETDISKSEQ:
 		return put_u64(argp, bdev->bd_disk->diskseq);
 	case BLKREPORTZONE:
+	case BLKREPORTZONEV2:
 		return blkdev_report_zones_ioctl(bdev, cmd, arg);
 	case BLKRESETZONE:
 	case BLKOPENZONE:

drivers/block/null_blk/null_blk.h

@@ -143,7 +143,8 @@ int null_init_zoned_dev(struct nullb_device *dev, struct queue_limits *lim);
 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);
+		      unsigned int nr_zones,
+		      struct blk_report_zones_args *args);
 blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_op op,
 				    sector_t sector, sector_t nr_sectors);
 size_t null_zone_valid_read_len(struct nullb *nullb,

drivers/block/null_blk/zoned.c

@@ -191,7 +191,7 @@ 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)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct nullb *nullb = disk->private_data;
 	struct nullb_device *dev = nullb->dev;
@@ -225,7 +225,7 @@ int null_report_zones(struct gendisk *disk, sector_t sector,
 		blkz.capacity = zone->capacity;
 		null_unlock_zone(dev, zone);
 
-		error = cb(&blkz, i, data);
+		error = disk_report_zone(disk, &blkz, i, args);
 		if (error)
 			return error;
 	}

drivers/block/ublk_drv.c

@@ -367,7 +367,7 @@ static void *ublk_alloc_report_buffer(struct ublk_device *ublk,
 }
 
 static int ublk_report_zones(struct gendisk *disk, sector_t sector,
-		      unsigned int nr_zones, report_zones_cb cb, void *data)
+		      unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct ublk_device *ub = disk->private_data;
 	unsigned int zone_size_sectors = disk->queue->limits.chunk_sectors;
@@ -430,7 +430,7 @@ static int ublk_report_zones(struct gendisk *disk, sector_t sector,
 			if (!zone->len)
 				break;
 
-			ret = cb(zone, i, data);
+			ret = disk_report_zone(disk, zone, i, args);
 			if (ret)
 				goto out;
 

drivers/block/virtio_blk.c

@@ -584,7 +584,8 @@ static int virtblk_submit_zone_report(struct virtio_blk *vblk,
 
 static int virtblk_parse_zone(struct virtio_blk *vblk,
 			       struct virtio_blk_zone_descriptor *entry,
-			       unsigned int idx, report_zones_cb cb, void *data)
+			       unsigned int idx,
+			       struct blk_report_zones_args *args)
 {
 	struct blk_zone zone = { };
 
@@ -650,12 +651,12 @@ static int virtblk_parse_zone(struct virtio_blk *vblk,
 	 * The callback below checks the validity of the reported
 	 * entry data, no need to further validate it here.
 	 */
-	return cb(&zone, idx, data);
+	return disk_report_zone(vblk->disk, &zone, idx, args);
 }
 
 static int virtblk_report_zones(struct gendisk *disk, sector_t sector,
-				 unsigned int nr_zones, report_zones_cb cb,
-				 void *data)
+				 unsigned int nr_zones,
+				 struct blk_report_zones_args *args)
 {
 	struct virtio_blk *vblk = disk->private_data;
 	struct virtio_blk_zone_report *report;
@@ -693,7 +694,7 @@ static int virtblk_report_zones(struct gendisk *disk, sector_t sector,
 
 		for (i = 0; i < nz && zone_idx < nr_zones; i++) {
 			ret = virtblk_parse_zone(vblk, &report->zones[i],
-						 zone_idx, cb, data);
+						 zone_idx, args);
 			if (ret)
 				goto fail_report;
 

drivers/block/zloop.c

@@ -647,7 +647,7 @@ static int zloop_open(struct gendisk *disk, blk_mode_t mode)
 }
 
 static int zloop_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct zloop_device *zlo = disk->private_data;
 	struct blk_zone blkz = {};
@@ -687,7 +687,7 @@ static int zloop_report_zones(struct gendisk *disk, sector_t sector,
 
 		mutex_unlock(&zone->lock);
 
-		ret = cb(&blkz, i, data);
+		ret = disk_report_zone(disk, &blkz, i, args);
 		if (ret)
 			return ret;
 	}

drivers/md/dm-zone.c

@@ -17,33 +17,26 @@
  * For internal zone reports bypassing the top BIO submission path.
  */
 static int dm_blk_do_report_zones(struct mapped_device *md, struct dm_table *t,
-				  sector_t sector, unsigned int nr_zones,
-				  report_zones_cb cb, void *data)
+				  unsigned int nr_zones,
+				  struct dm_report_zones_args *args)
 {
-	struct gendisk *disk = md->disk;
-	int ret;
-	struct dm_report_zones_args args = {
-		.next_sector = sector,
-		.orig_data = data,
-		.orig_cb = cb,
-	};
-
 	do {
 		struct dm_target *tgt;
+		int ret;
 
-		tgt = dm_table_find_target(t, args.next_sector);
+		tgt = dm_table_find_target(t, args->next_sector);
 		if (WARN_ON_ONCE(!tgt->type->report_zones))
 			return -EIO;
 
-		args.tgt = tgt;
-		ret = tgt->type->report_zones(tgt, &args,
-					      nr_zones - args.zone_idx);
+		args->tgt = tgt;
+		ret = tgt->type->report_zones(tgt, args,
+					      nr_zones - args->zone_idx);
 		if (ret < 0)
 			return ret;
-	} while (args.zone_idx < nr_zones &&
-		 args.next_sector < get_capacity(disk));
+	} while (args->zone_idx < nr_zones &&
+		 args->next_sector < get_capacity(md->disk));
 
-	return args.zone_idx;
+	return args->zone_idx;
 }
 
 /*
@@ -52,7 +45,8 @@ static int dm_blk_do_report_zones(struct mapped_device *md, struct dm_table *t,
  * generally implemented by targets using dm_report_zones().
  */
 int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data)
+			unsigned int nr_zones,
+			struct blk_report_zones_args *args)
 {
 	struct mapped_device *md = disk->private_data;
 	struct dm_table *map;
@@ -76,9 +70,14 @@ int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
 		map = zone_revalidate_map;
 	}
 
-	if (map)
-		ret = dm_blk_do_report_zones(md, map, sector, nr_zones, cb,
-					     data);
+	if (map) {
+		struct dm_report_zones_args dm_args = {
+			.disk = md->disk,
+			.next_sector = sector,
+			.rep_args = args,
+		};
+		ret = dm_blk_do_report_zones(md, map, nr_zones, &dm_args);
+	}
 
 	if (put_table)
 		dm_put_live_table(md, srcu_idx);
@@ -113,7 +112,9 @@ static int dm_report_zones_cb(struct blk_zone *zone, unsigned int idx,
 	}
 
 	args->next_sector = zone->start + zone->len;
-	return args->orig_cb(zone, args->zone_idx++, args->orig_data);
+
+	return disk_report_zone(args->disk, zone, args->zone_idx++,
+				args->rep_args);
 }
 
 /*
@@ -492,10 +493,15 @@ int dm_zone_get_reset_bitmap(struct mapped_device *md, struct dm_table *t,
 			     sector_t sector, unsigned int nr_zones,
 			     unsigned long *need_reset)
 {
+	struct dm_report_zones_args args = {
+		.disk = md->disk,
+		.next_sector = sector,
+		.cb = dm_zone_need_reset_cb,
+		.data = need_reset,
+	};
 	int ret;
 
-	ret = dm_blk_do_report_zones(md, t, sector, nr_zones,
-				     dm_zone_need_reset_cb, need_reset);
+	ret = dm_blk_do_report_zones(md, t, nr_zones, &args);
 	if (ret != nr_zones) {
 		DMERR("Get %s zone reset bitmap failed\n",
 		      md->disk->disk_name);

drivers/md/dm.h

@@ -109,7 +109,8 @@ void dm_finalize_zone_settings(struct dm_table *t, struct queue_limits *lim);
 void dm_zone_endio(struct dm_io *io, struct bio *clone);
 #ifdef CONFIG_BLK_DEV_ZONED
 int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data);
+			unsigned int nr_zones,
+			struct blk_report_zones_args *args);
 bool dm_is_zone_write(struct mapped_device *md, struct bio *bio);
 int dm_zone_get_reset_bitmap(struct mapped_device *md, struct dm_table *t,
 			     sector_t sector, unsigned int nr_zones,

drivers/nvme/host/core.c

@@ -2599,10 +2599,9 @@ static void nvme_configure_opal(struct nvme_ctrl *ctrl, bool was_suspended)
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static int nvme_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
-	return nvme_ns_report_zones(disk->private_data, sector, nr_zones, cb,
-			data);
+	return nvme_ns_report_zones(disk->private_data, sector, nr_zones, args);
 }
 #else
 #define nvme_report_zones	NULL

drivers/nvme/host/multipath.c

@@ -576,7 +576,7 @@ static int nvme_ns_head_get_unique_id(struct gendisk *disk, u8 id[16],
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct nvme_ns_head *head = disk->private_data;
 	struct nvme_ns *ns;
@@ -585,7 +585,7 @@ static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector,
 	srcu_idx = srcu_read_lock(&head->srcu);
 	ns = nvme_find_path(head);
 	if (ns)
-		ret = nvme_ns_report_zones(ns, sector, nr_zones, cb, data);
+		ret = nvme_ns_report_zones(ns, sector, nr_zones, args);
 	srcu_read_unlock(&head->srcu, srcu_idx);
 	return ret;
 }

drivers/nvme/host/nvme.h

@@ -1108,7 +1108,7 @@ struct nvme_zone_info {
 };
 
 int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data);
+		unsigned int nr_zones, struct blk_report_zones_args *args);
 int nvme_query_zone_info(struct nvme_ns *ns, unsigned lbaf,
 		struct nvme_zone_info *zi);
 void nvme_update_zone_info(struct nvme_ns *ns, struct queue_limits *lim,

drivers/nvme/host/zns.c

@@ -148,8 +148,8 @@ static void *nvme_zns_alloc_report_buffer(struct nvme_ns *ns,
 
 static int nvme_zone_parse_entry(struct nvme_ns *ns,
 				 struct nvme_zone_descriptor *entry,
-				 unsigned int idx, report_zones_cb cb,
-				 void *data)
+				 unsigned int idx,
+				 struct blk_report_zones_args *args)
 {
 	struct nvme_ns_head *head = ns->head;
 	struct blk_zone zone = { };
@@ -169,11 +169,11 @@ static int nvme_zone_parse_entry(struct nvme_ns *ns,
 	else
 		zone.wp = nvme_lba_to_sect(head, le64_to_cpu(entry->wp));
 
-	return cb(&zone, idx, data);
+	return disk_report_zone(ns->disk, &zone, idx, args);
 }
 
 int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data)
+		unsigned int nr_zones, struct blk_report_zones_args *args)
 {
 	struct nvme_zone_report *report;
 	struct nvme_command c = { };
@@ -213,7 +213,7 @@ int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector,
 
 		for (i = 0; i < nz && zone_idx < nr_zones; i++) {
 			ret = nvme_zone_parse_entry(ns, &report->entries[i],
-						    zone_idx, cb, data);
+						    zone_idx, args);
 			if (ret)
 				goto out_free;
 			zone_idx++;

drivers/scsi/sd.h

@@ -240,7 +240,7 @@ blk_status_t sd_zbc_setup_zone_mgmt_cmnd(struct scsi_cmnd *cmd,
 unsigned int sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes,
 			     struct scsi_sense_hdr *sshdr);
 int sd_zbc_report_zones(struct gendisk *disk, sector_t sector,
-		unsigned int nr_zones, report_zones_cb cb, void *data);
+		unsigned int nr_zones, struct blk_report_zones_args *args);
 
 #else /* CONFIG_BLK_DEV_ZONED */
 

drivers/scsi/sd_zbc.c

@@ -35,8 +35,7 @@ static bool sd_zbc_is_gap_zone(const u8 buf[64])
  * @buf: SCSI zone descriptor.
  * @idx: Index of the zone relative to the first zone reported by the current
  *	sd_zbc_report_zones() call.
- * @cb: Callback function pointer.
- * @data: Second argument passed to @cb.
+ * @args: report zones arguments (callback, etc)
  *
  * Return: Value returned by @cb.
  *
@@ -44,12 +43,11 @@ static bool sd_zbc_is_gap_zone(const u8 buf[64])
  * call @cb(blk_zone, @data).
  */
 static int sd_zbc_parse_report(struct scsi_disk *sdkp, const u8 buf[64],
-			       unsigned int idx, report_zones_cb cb, void *data)
+			unsigned int idx, struct blk_report_zones_args *args)
 {
 	struct scsi_device *sdp = sdkp->device;
 	struct blk_zone zone = { 0 };
 	sector_t start_lba, gran;
-	int ret;
 
 	if (WARN_ON_ONCE(sd_zbc_is_gap_zone(buf)))
 		return -EINVAL;
@@ -87,11 +85,7 @@ static int sd_zbc_parse_report(struct scsi_disk *sdkp, const u8 buf[64],
 	else
 		zone.wp = logical_to_sectors(sdp, get_unaligned_be64(&buf[24]));
 
-	ret = cb(&zone, idx, data);
-	if (ret)
-		return ret;
-
-	return 0;
+	return disk_report_zone(sdkp->disk, &zone, idx, args);
 }
 
 /**
@@ -217,14 +211,14 @@ static inline sector_t sd_zbc_zone_sectors(struct scsi_disk *sdkp)
  * @disk: Disk to report zones for.
  * @sector: Start sector.
  * @nr_zones: Maximum number of zones to report.
- * @cb: Callback function called to report zone information.
- * @data: Second argument passed to @cb.
+ * @args: Callback arguments.
  *
  * Called by the block layer to iterate over zone information. See also the
  * disk->fops->report_zones() calls in block/blk-zoned.c.
  */
 int sd_zbc_report_zones(struct gendisk *disk, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data)
+			unsigned int nr_zones,
+			struct blk_report_zones_args *args)
 {
 	struct scsi_disk *sdkp = scsi_disk(disk);
 	sector_t lba = sectors_to_logical(sdkp->device, sector);
@@ -283,7 +277,7 @@ int sd_zbc_report_zones(struct gendisk *disk, sector_t sector,
 			}
 
 			ret = sd_zbc_parse_report(sdkp, buf + offset, zone_idx,
-						  cb, data);
+						  args);
 			if (ret)
 				goto out;
 

fs/btrfs/zoned.c

@@ -264,8 +264,8 @@ static int btrfs_get_dev_zones(struct btrfs_device *device, u64 pos,
 		}
 	}
 
-	ret = blkdev_report_zones(device->bdev, pos >> SECTOR_SHIFT, *nr_zones,
-				  copy_zone_info_cb, zones);
+	ret = blkdev_report_zones_cached(device->bdev, pos >> SECTOR_SHIFT,
+					 *nr_zones, copy_zone_info_cb, zones);
 	if (ret < 0) {
 		btrfs_err(device->fs_info,
 				 "zoned: failed to read zone %llu on %s (devid %llu)",
@@ -494,6 +494,7 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
 			case BLK_ZONE_COND_IMP_OPEN:
 			case BLK_ZONE_COND_EXP_OPEN:
 			case BLK_ZONE_COND_CLOSED:
+			case BLK_ZONE_COND_ACTIVE:
 				__set_bit(nreported, zone_info->active_zones);
 				nactive++;
 				break;
@@ -896,9 +897,9 @@ int btrfs_sb_log_location_bdev(struct block_device *bdev, int mirror, int rw,
 	if (sb_zone + 1 >= nr_zones)
 		return -ENOENT;
 
-	ret = blkdev_report_zones(bdev, zone_start_sector(sb_zone, bdev),
-				  BTRFS_NR_SB_LOG_ZONES, copy_zone_info_cb,
-				  zones);
+	ret = blkdev_report_zones_cached(bdev, zone_start_sector(sb_zone, bdev),
+					 BTRFS_NR_SB_LOG_ZONES,
+					 copy_zone_info_cb, zones);
 	if (ret < 0)
 		return ret;
 	if (unlikely(ret != BTRFS_NR_SB_LOG_ZONES))

fs/xfs/libxfs/xfs_zones.c

@@ -95,6 +95,7 @@ xfs_zone_validate_seq(
 	case BLK_ZONE_COND_IMP_OPEN:
 	case BLK_ZONE_COND_EXP_OPEN:
 	case BLK_ZONE_COND_CLOSED:
+	case BLK_ZONE_COND_ACTIVE:
 		return xfs_zone_validate_wp(zone, rtg, write_pointer);
 	case BLK_ZONE_COND_FULL:
 		return xfs_zone_validate_full(zone, rtg, write_pointer);

fs/xfs/xfs_zone_alloc.c

@@ -1250,7 +1250,7 @@ xfs_mount_zones(
 	trace_xfs_zones_mount(mp);
 
 	if (bdev_is_zoned(bt->bt_bdev)) {
-		error = blkdev_report_zones(bt->bt_bdev,
+		error = blkdev_report_zones_cached(bt->bt_bdev,
 				XFS_FSB_TO_BB(mp, mp->m_sb.sb_rtstart),
 				mp->m_sb.sb_rgcount, xfs_get_zone_info_cb, &iz);
 		if (error < 0)

include/linux/blkdev.h

@@ -38,6 +38,7 @@ struct blk_flush_queue;
 struct kiocb;
 struct pr_ops;
 struct rq_qos;
+struct blk_report_zones_args;
 struct blk_queue_stats;
 struct blk_stat_callback;
 struct blk_crypto_profile;
@@ -195,7 +196,7 @@ struct gendisk {
 	unsigned int		nr_zones;
 	unsigned int		zone_capacity;
 	unsigned int		last_zone_capacity;
-	unsigned long __rcu	*conv_zones_bitmap;
+	u8 __rcu		*zones_cond;
 	unsigned int		zone_wplugs_hash_bits;
 	atomic_t		nr_zone_wplugs;
 	spinlock_t		zone_wplugs_lock;
@@ -432,9 +433,17 @@ struct queue_limits {
 typedef int (*report_zones_cb)(struct blk_zone *zone, unsigned int idx,
 			       void *data);
 
+int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
+		     unsigned int idx, struct blk_report_zones_args *args);
+
+int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
+			 struct blk_zone *zone);
+
 #define BLK_ALL_ZONES  ((unsigned int)-1)
 int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 		unsigned int nr_zones, report_zones_cb cb, void *data);
+int blkdev_report_zones_cached(struct block_device *bdev, sector_t sector,
+		unsigned int nr_zones, report_zones_cb cb, void *data);
 int blkdev_zone_mgmt(struct block_device *bdev, enum req_op op,
 		sector_t sectors, sector_t nr_sectors);
 int blk_revalidate_disk_zones(struct gendisk *disk);
@@ -921,12 +930,20 @@ static inline unsigned int bdev_zone_capacity(struct block_device *bdev,
 {
 	return disk_zone_capacity(bdev->bd_disk, pos);
 }
+
+bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector);
+
 #else /* CONFIG_BLK_DEV_ZONED */
 static inline unsigned int disk_nr_zones(struct gendisk *disk)
 {
 	return 0;
 }
 
+static inline bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector)
+{
+	return false;
+}
+
 static inline bool bio_needs_zone_write_plugging(struct bio *bio)
 {
 	return false;
@@ -1529,33 +1546,6 @@ static inline bool bdev_is_zone_aligned(struct block_device *bdev,
 	return bdev_is_zone_start(bdev, sector);
 }
 
-/**
- * bdev_zone_is_seq - check if a sector belongs to a sequential write zone
- * @bdev:	block device to check
- * @sector:	sector number
- *
- * Check if @sector on @bdev is contained in a sequential write required zone.
- */
-static inline bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector)
-{
-	bool is_seq = false;
-
-#if IS_ENABLED(CONFIG_BLK_DEV_ZONED)
-	if (bdev_is_zoned(bdev)) {
-		struct gendisk *disk = bdev->bd_disk;
-		unsigned long *bitmap;
-
-		rcu_read_lock();
-		bitmap = rcu_dereference(disk->conv_zones_bitmap);
-		is_seq = !bitmap ||
-			!test_bit(disk_zone_no(disk, sector), bitmap);
-		rcu_read_unlock();
-	}
-#endif
-
-	return is_seq;
-}
-
 int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
 			   sector_t nr_sects, gfp_t gfp_mask);
 
@@ -1662,7 +1652,8 @@ struct block_device_operations {
 	/* this callback is with swap_lock and sometimes page table lock held */
 	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
 	int (*report_zones)(struct gendisk *, sector_t sector,
-			unsigned int nr_zones, report_zones_cb cb, void *data);
+			    unsigned int nr_zones,
+			    struct blk_report_zones_args *args);
 	char *(*devnode)(struct gendisk *disk, umode_t *mode);
 	/* returns the length of the identifier or a negative errno: */
 	int (*get_unique_id)(struct gendisk *disk, u8 id[16],

include/linux/device-mapper.h

@@ -538,12 +538,18 @@ void dm_submit_bio_remap(struct bio *clone, struct bio *tgt_clone);
 #ifdef CONFIG_BLK_DEV_ZONED
 struct dm_report_zones_args {
 	struct dm_target *tgt;
+	struct gendisk *disk;
 	sector_t next_sector;
 
-	void *orig_data;
-	report_zones_cb orig_cb;
 	unsigned int zone_idx;
 
+	/* for block layer ->report_zones */
+	struct blk_report_zones_args *rep_args;
+
+	/* for internal users */
+	report_zones_cb cb;
+	void *data;
+
 	/* must be filled by ->report_zones before calling dm_report_zones_cb */
 	sector_t start;
 };

include/uapi/linux/blkzoned.h

@@ -48,6 +48,8 @@ enum blk_zone_type {
  *                      FINISH ZONE command.
  * @BLK_ZONE_COND_READONLY: The zone is read-only.
  * @BLK_ZONE_COND_OFFLINE: The zone is offline (sectors cannot be read/written).
+ * @BLK_ZONE_COND_ACTIVE: The zone is either implicitly open, explicitly open,
+ *			  or closed.
  *
  * The Zone Condition state machine in the ZBC/ZAC standards maps the above
  * deinitions as:
@@ -61,6 +63,13 @@ enum blk_zone_type {
  *
  * Conditions 0x5 to 0xC are reserved by the current ZBC/ZAC spec and should
  * be considered invalid.
+ *
+ * The condition BLK_ZONE_COND_ACTIVE is used only with cached zone reports.
+ * It is used to report any of the BLK_ZONE_COND_IMP_OPEN,
+ * BLK_ZONE_COND_EXP_OPEN and BLK_ZONE_COND_CLOSED conditions. Conversely, a
+ * regular zone report will never report a zone condition using
+ * BLK_ZONE_COND_ACTIVE and instead use the conditions BLK_ZONE_COND_IMP_OPEN,
+ * BLK_ZONE_COND_EXP_OPEN or BLK_ZONE_COND_CLOSED as reported by the device.
  */
 enum blk_zone_cond {
 	BLK_ZONE_COND_NOT_WP	= 0x0,
@@ -71,15 +80,27 @@ enum blk_zone_cond {
 	BLK_ZONE_COND_READONLY	= 0xD,
 	BLK_ZONE_COND_FULL	= 0xE,
 	BLK_ZONE_COND_OFFLINE	= 0xF,
+
+	BLK_ZONE_COND_ACTIVE	= 0xFF,
 };
 
 /**
  * enum blk_zone_report_flags - Feature flags of reported zone descriptors.
  *
- * @BLK_ZONE_REP_CAPACITY: Zone descriptor has capacity field.
+ * @BLK_ZONE_REP_CAPACITY: Output only. Indicates that zone descriptors in a
+ *			   zone report have a valid capacity field.
+ * @BLK_ZONE_REP_CACHED: Input only. Indicates that the zone report should be
+ *			 generated using cached zone information. In this case,
+ *			 the implicit open, explicit open and closed zone
+ *			 conditions are all reported with the
+ *			 BLK_ZONE_COND_ACTIVE condition.
  */
 enum blk_zone_report_flags {
-	BLK_ZONE_REP_CAPACITY	= (1 << 0),
+	/* Output flags */
+	BLK_ZONE_REP_CAPACITY	= (1U << 0),
+
+	/* Input flags */
+	BLK_ZONE_REP_CACHED	= (1U << 31),
 };
 
 /**
@@ -122,6 +143,10 @@ struct blk_zone {
  * @sector: starting sector of report
  * @nr_zones: IN maximum / OUT actual
  * @flags: one or more flags as defined by enum blk_zone_report_flags.
+ * @flags: one or more flags as defined by enum blk_zone_report_flags.
+ *	   With BLKREPORTZONE, this field is ignored as an input and is valid
+ *	   only as an output. Using BLKREPORTZONEV2, this field is used as both
+ *	   input and output.
  * @zones: Space to hold @nr_zones @zones entries on reply.
  *
  * The array of at most @nr_zones must follow this structure in memory.
@@ -148,9 +173,19 @@ struct blk_zone_range {
 /**
  * Zoned block device ioctl's:
  *
- * @BLKREPORTZONE: Get zone information. Takes a zone report as argument.
- *                 The zone report will start from the zone containing the
- *                 sector specified in the report request structure.
+ * @BLKREPORTZONE: Get zone information from a zoned device. Takes a zone report
+ *		   as argument. The zone report will start from the zone
+ *		   containing the sector specified in struct blk_zone_report.
+ *		   The flags field of struct blk_zone_report is used as an
+ *		   output only and ignored as an input.
+ *		   DEPRECATED, use BLKREPORTZONEV2 instead.
+ * @BLKREPORTZONEV2: Same as @BLKREPORTZONE but uses the flags field of
+ *		     struct blk_zone_report as an input, allowing to get a zone
+ *		     report using cached zone information if the flag
+ *		     BLK_ZONE_REP_CACHED is set. In such case, the zone report
+ *		     may include zones with the condition @BLK_ZONE_COND_ACTIVE
+ *		     (c.f. the description of this condition above for more
+ *		     details).
  * @BLKRESETZONE: Reset the write pointer of the zones in the specified
  *                sector range. The sector range must be zone aligned.
  * @BLKGETZONESZ: Get the device zone size in number of 512 B sectors.
@@ -169,5 +204,6 @@ struct blk_zone_range {
 #define BLKOPENZONE	_IOW(0x12, 134, struct blk_zone_range)
 #define BLKCLOSEZONE	_IOW(0x12, 135, struct blk_zone_range)
 #define BLKFINISHZONE	_IOW(0x12, 136, struct blk_zone_range)
+#define BLKREPORTZONEV2	_IOWR(0x12, 142, struct blk_zone_report)
 
 #endif /* _UAPI_BLKZONED_H */

include/uapi/linux/fs.h

@@ -298,7 +298,7 @@ struct file_attr {
 #define BLKROTATIONAL _IO(0x12,126)
 #define BLKZEROOUT _IO(0x12,127)
 #define BLKGETDISKSEQ _IOR(0x12,128,__u64)
-/* 130-136 are used by zoned block device ioctls (uapi/linux/blkzoned.h) */
+/* 130-136 and 142 are used by zoned block device ioctls (uapi/linux/blkzoned.h) */
 /* 137-141 are used by blk-crypto ioctls (uapi/linux/blk-crypto.h) */
 #define BLKTRACESETUP2 _IOWR(0x12, 142, struct blk_user_trace_setup2)