summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--block/blk-zoned.c798
-rw-r--r--block/blk.h14
-rw-r--r--block/ioctl.c1
-rw-r--r--drivers/block/null_blk/null_blk.h3
-rw-r--r--drivers/block/null_blk/zoned.c4
-rw-r--r--drivers/block/ublk_drv.c4
-rw-r--r--drivers/block/virtio_blk.c11
-rw-r--r--drivers/block/zloop.c4
-rw-r--r--drivers/md/dm-zone.c54
-rw-r--r--drivers/md/dm.h3
-rw-r--r--drivers/nvme/host/core.c5
-rw-r--r--drivers/nvme/host/multipath.c4
-rw-r--r--drivers/nvme/host/nvme.h2
-rw-r--r--drivers/nvme/host/zns.c10
-rw-r--r--drivers/scsi/sd.h2
-rw-r--r--drivers/scsi/sd_zbc.c20
-rw-r--r--fs/btrfs/zoned.c11
-rw-r--r--fs/xfs/libxfs/xfs_zones.c1
-rw-r--r--fs/xfs/xfs_zone_alloc.c2
-rw-r--r--include/linux/blkdev.h49
-rw-r--r--include/linux/device-mapper.h10
-rw-r--r--include/uapi/linux/blkzoned.h46
-rw-r--r--include/uapi/linux/fs.h2
23 files changed, 784 insertions, 276 deletions
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 5e2a5788dc3b..bba64b427082 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -33,12 +33,18 @@ static const char *const zone_cond_name[] = {
ZONE_COND_NAME(READONLY),
ZONE_COND_NAME(FULL),
ZONE_COND_NAME(OFFLINE),
+ ZONE_COND_NAME(ACTIVE),
};
#undef ZONE_COND_NAME
/*
* Per-zone write plug.
* @node: hlist_node structure for managing the plug using a hash table.
+ * @bio_list: The list of BIOs that are currently plugged.
+ * @bio_work: Work struct to handle issuing of plugged BIOs
+ * @rcu_head: RCU head to free zone write plugs with an RCU grace period.
+ * @disk: The gendisk the plug belongs to.
+ * @lock: Spinlock to atomically manipulate the plug.
* @ref: Zone write plug reference counter. A zone write plug reference is
* always at least 1 when the plug is hashed in the disk plug hash table.
* The reference is incremented whenever a new BIO needing plugging is
@@ -48,29 +54,44 @@ static const char *const zone_cond_name[] = {
* reference is dropped whenever the zone of the zone write plug is reset,
* finished and when the zone becomes full (last write BIO to the zone
* completes).
- * @lock: Spinlock to atomically manipulate the plug.
* @flags: Flags indicating the plug state.
* @zone_no: The number of the zone the plug is managing.
* @wp_offset: The zone write pointer location relative to the start of the zone
* as a number of 512B sectors.
- * @bio_list: The list of BIOs that are currently plugged.
- * @bio_work: Work struct to handle issuing of plugged BIOs
- * @rcu_head: RCU head to free zone write plugs with an RCU grace period.
- * @disk: The gendisk the plug belongs to.
+ * @cond: Condition of the zone
*/
struct blk_zone_wplug {
struct hlist_node node;
- refcount_t ref;
- spinlock_t lock;
- unsigned int flags;
- unsigned int zone_no;
- unsigned int wp_offset;
struct bio_list bio_list;
struct work_struct bio_work;
struct rcu_head rcu_head;
struct gendisk *disk;
+ spinlock_t lock;
+ refcount_t ref;
+ unsigned int flags;
+ unsigned int zone_no;
+ unsigned int wp_offset;
+ enum blk_zone_cond cond;
};
+static inline bool disk_need_zone_resources(struct gendisk *disk)
+{
+ /*
+ * All request-based zoned devices need zone resources so that the
+ * block layer can automatically handle write BIO plugging. BIO-based
+ * device drivers (e.g. DM devices) are normally responsible for
+ * handling zone write ordering and do not need zone resources, unless
+ * the driver requires zone append emulation.
+ */
+ return queue_is_mq(disk->queue) ||
+ queue_emulates_zone_append(disk->queue);
+}
+
+static inline unsigned int disk_zone_wplugs_hash_size(struct gendisk *disk)
+{
+ return 1U << disk->zone_wplugs_hash_bits;
+}
+
/*
* Zone write plug flags bits:
* - BLK_ZONE_WPLUG_PLUGGED: Indicates that the zone write plug is plugged,
@@ -109,28 +130,108 @@ const char *blk_zone_cond_str(enum blk_zone_cond zone_cond)
}
EXPORT_SYMBOL_GPL(blk_zone_cond_str);
-struct disk_report_zones_cb_args {
- struct gendisk *disk;
- report_zones_cb user_cb;
- void *user_data;
-};
+static void blk_zone_set_cond(u8 *zones_cond, unsigned int zno,
+ enum blk_zone_cond cond)
+{
+ if (!zones_cond)
+ return;
-static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
- struct blk_zone *zone);
+ switch (cond) {
+ case BLK_ZONE_COND_IMP_OPEN:
+ case BLK_ZONE_COND_EXP_OPEN:
+ case BLK_ZONE_COND_CLOSED:
+ zones_cond[zno] = BLK_ZONE_COND_ACTIVE;
+ return;
+ case BLK_ZONE_COND_NOT_WP:
+ case BLK_ZONE_COND_EMPTY:
+ case BLK_ZONE_COND_FULL:
+ case BLK_ZONE_COND_OFFLINE:
+ case BLK_ZONE_COND_READONLY:
+ default:
+ zones_cond[zno] = cond;
+ return;
+ }
+}
-static int disk_report_zones_cb(struct blk_zone *zone, unsigned int idx,
- void *data)
+static void disk_zone_set_cond(struct gendisk *disk, sector_t sector,
+ enum blk_zone_cond cond)
{
- struct disk_report_zones_cb_args *args = data;
- struct gendisk *disk = args->disk;
+ u8 *zones_cond;
- if (disk->zone_wplugs_hash)
- disk_zone_wplug_sync_wp_offset(disk, zone);
+ rcu_read_lock();
+ zones_cond = rcu_dereference(disk->zones_cond);
+ if (zones_cond) {
+ unsigned int zno = disk_zone_no(disk, sector);
+
+ /*
+ * The condition of a conventional, readonly and offline zones
+ * never changes, so do nothing if the target zone is in one of
+ * these conditions.
+ */
+ switch (zones_cond[zno]) {
+ case BLK_ZONE_COND_NOT_WP:
+ case BLK_ZONE_COND_READONLY:
+ case BLK_ZONE_COND_OFFLINE:
+ break;
+ default:
+ blk_zone_set_cond(zones_cond, zno, cond);
+ break;
+ }
+ }
+ rcu_read_unlock();
+}
+
+/**
+ * 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.
+ */
+bool bdev_zone_is_seq(struct block_device *bdev, sector_t sector)
+{
+ struct gendisk *disk = bdev->bd_disk;
+ unsigned int zno = disk_zone_no(disk, sector);
+ bool is_seq = false;
+ u8 *zones_cond;
+
+ if (!bdev_is_zoned(bdev))
+ return false;
+
+ rcu_read_lock();
+ zones_cond = rcu_dereference(disk->zones_cond);
+ if (zones_cond && zno < disk->nr_zones)
+ is_seq = zones_cond[zno] != BLK_ZONE_COND_NOT_WP;
+ rcu_read_unlock();
+
+ return is_seq;
+}
+EXPORT_SYMBOL_GPL(bdev_zone_is_seq);
+
+/*
+ * Zone report arguments for block device drivers report_zones operation.
+ * @cb: report_zones_cb callback for each reported zone.
+ * @data: Private data passed to report_zones_cb.
+ */
+struct blk_report_zones_args {
+ report_zones_cb cb;
+ void *data;
+ bool report_active;
+};
+
+static int blkdev_do_report_zones(struct block_device *bdev, sector_t sector,
+ unsigned int nr_zones,
+ struct blk_report_zones_args *args)
+{
+ struct gendisk *disk = bdev->bd_disk;
+
+ if (!bdev_is_zoned(bdev) || WARN_ON_ONCE(!disk->fops->report_zones))
+ return -EOPNOTSUPP;
- if (!args->user_cb)
+ if (!nr_zones || sector >= get_capacity(disk))
return 0;
- return args->user_cb(zone, idx, args->user_data);
+ return disk->fops->report_zones(disk, sector, nr_zones, args);
}
/**
@@ -155,22 +256,12 @@ static int disk_report_zones_cb(struct blk_zone *zone, unsigned int idx,
int blkdev_report_zones(struct block_device *bdev, sector_t sector,
unsigned int nr_zones, report_zones_cb cb, void *data)
{
- struct gendisk *disk = bdev->bd_disk;
- sector_t capacity = get_capacity(disk);
- struct disk_report_zones_cb_args args = {
- .disk = disk,
- .user_cb = cb,
- .user_data = data,
+ struct blk_report_zones_args args = {
+ .cb = cb,
+ .data = data,
};
- if (!bdev_is_zoned(bdev) || WARN_ON_ONCE(!disk->fops->report_zones))
- return -EOPNOTSUPP;
-
- if (!nr_zones || sector >= capacity)
- return 0;
-
- return disk->fops->report_zones(disk, sector, nr_zones,
- disk_report_zones_cb, &args);
+ return blkdev_do_report_zones(bdev, sector, nr_zones, &args);
}
EXPORT_SYMBOL_GPL(blkdev_report_zones);
@@ -266,7 +357,12 @@ static int blkdev_copy_zone_to_user(struct blk_zone *zone, unsigned int idx,
}
/*
- * BLKREPORTZONE ioctl processing.
+ * Mask of valid input flags for BLKREPORTZONEV2 ioctl.
+ */
+#define BLK_ZONE_REPV2_INPUT_FLAGS BLK_ZONE_REP_CACHED
+
+/*
+ * BLKREPORTZONE and BLKREPORTZONEV2 ioctl processing.
* Called from blkdev_ioctl.
*/
int blkdev_report_zones_ioctl(struct block_device *bdev, unsigned int cmd,
@@ -290,8 +386,22 @@ int blkdev_report_zones_ioctl(struct block_device *bdev, unsigned int cmd,
return -EINVAL;
args.zones = argp + sizeof(struct blk_zone_report);
- ret = blkdev_report_zones(bdev, rep.sector, rep.nr_zones,
- blkdev_copy_zone_to_user, &args);
+
+ switch (cmd) {
+ case BLKREPORTZONE:
+ ret = blkdev_report_zones(bdev, rep.sector, rep.nr_zones,
+ blkdev_copy_zone_to_user, &args);
+ break;
+ case BLKREPORTZONEV2:
+ if (rep.flags & ~BLK_ZONE_REPV2_INPUT_FLAGS)
+ return -EINVAL;
+ ret = blkdev_report_zones_cached(bdev, rep.sector, rep.nr_zones,
+ blkdev_copy_zone_to_user, &args);
+ break;
+ default:
+ return -EINVAL;
+ }
+
if (ret < 0)
return ret;
@@ -401,6 +511,7 @@ static bool disk_insert_zone_wplug(struct gendisk *disk,
{
struct blk_zone_wplug *zwplg;
unsigned long flags;
+ u8 *zones_cond;
unsigned int idx =
hash_32(zwplug->zone_no, disk->zone_wplugs_hash_bits);
@@ -416,6 +527,20 @@ static bool disk_insert_zone_wplug(struct gendisk *disk,
return false;
}
}
+
+ /*
+ * Set the zone condition: if we do not yet have a zones_cond array
+ * attached to the disk, then this is a zone write plug insert from the
+ * first call to blk_revalidate_disk_zones(), in which case the zone is
+ * necessarilly in the active condition.
+ */
+ zones_cond = rcu_dereference_check(disk->zones_cond,
+ lockdep_is_held(&disk->zone_wplugs_lock));
+ if (zones_cond)
+ zwplug->cond = zones_cond[zwplug->zone_no];
+ else
+ zwplug->cond = BLK_ZONE_COND_ACTIVE;
+
hlist_add_head_rcu(&zwplug->node, &disk->zone_wplugs_hash[idx]);
atomic_inc(&disk->nr_zone_wplugs);
spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
@@ -515,10 +640,15 @@ static void disk_remove_zone_wplug(struct gendisk *disk,
/*
* Mark the zone write plug as unhashed and drop the extra reference we
- * took when the plug was inserted in the hash table.
+ * took when the plug was inserted in the hash table. Also update the
+ * disk zone condition array with the current condition of the zone
+ * write plug.
*/
zwplug->flags |= BLK_ZONE_WPLUG_UNHASHED;
spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
+ blk_zone_set_cond(rcu_dereference_check(disk->zones_cond,
+ lockdep_is_held(&disk->zone_wplugs_lock)),
+ zwplug->zone_no, zwplug->cond);
hlist_del_init_rcu(&zwplug->node);
atomic_dec(&disk->nr_zone_wplugs);
spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
@@ -621,6 +751,22 @@ static void disk_zone_wplug_abort(struct blk_zone_wplug *zwplug)
}
/*
+ * Update a zone write plug condition based on the write pointer offset.
+ */
+static void disk_zone_wplug_update_cond(struct gendisk *disk,
+ struct blk_zone_wplug *zwplug)
+{
+ lockdep_assert_held(&zwplug->lock);
+
+ if (disk_zone_wplug_is_full(disk, zwplug))
+ zwplug->cond = BLK_ZONE_COND_FULL;
+ else if (!zwplug->wp_offset)
+ zwplug->cond = BLK_ZONE_COND_EMPTY;
+ else
+ zwplug->cond = BLK_ZONE_COND_ACTIVE;
+}
+
+/*
* Set a zone write plug write pointer offset to the specified value.
* This aborts all plugged BIOs, which is fine as this function is called for
* a zone reset operation, a zone finish operation or if the zone needs a wp
@@ -635,6 +781,8 @@ static void disk_zone_wplug_set_wp_offset(struct gendisk *disk,
/* Update the zone write pointer and abort all plugged BIOs. */
zwplug->flags &= ~BLK_ZONE_WPLUG_NEED_WP_UPDATE;
zwplug->wp_offset = wp_offset;
+ disk_zone_wplug_update_cond(disk, zwplug);
+
disk_zone_wplug_abort(zwplug);
/*
@@ -688,81 +836,333 @@ static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
disk_put_zone_wplug(zwplug);
}
-static int disk_zone_sync_wp_offset(struct gendisk *disk, sector_t sector)
+/**
+ * disk_report_zone - Report one zone
+ * @disk: Target disk
+ * @zone: The zone to report
+ * @idx: The index of the zone in the overall zone report
+ * @args: report zones callback and data
+ *
+ * Description:
+ * Helper function for block device drivers to report one zone of a zone
+ * report initiated with blkdev_report_zones(). The zone being reported is
+ * specified by @zone and used to update, if necessary, the zone write plug
+ * information for the zone. If @args specifies a user callback function,
+ * this callback is executed.
+ */
+int disk_report_zone(struct gendisk *disk, struct blk_zone *zone,
+ unsigned int idx, struct blk_report_zones_args *args)
+{
+ if (args->report_active) {
+ /*
+ * If we come here, then this is a report zones as a fallback
+ * for a cached report. So collapse the implicit open, explicit
+ * open and closed conditions into the active zone condition.
+ */
+ switch (zone->cond) {
+ case BLK_ZONE_COND_IMP_OPEN:
+ case BLK_ZONE_COND_EXP_OPEN:
+ case BLK_ZONE_COND_CLOSED:
+ zone->cond = BLK_ZONE_COND_ACTIVE;
+ break;
+ default:
+ break;
+ }
+ }
+
+ if (disk->zone_wplugs_hash)
+ disk_zone_wplug_sync_wp_offset(disk, zone);
+
+ if (args && args->cb)
+ return args->cb(zone, idx, args->data);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(disk_report_zone);
+
+static int blkdev_report_zone_cb(struct blk_zone *zone, unsigned int idx,
+ void *data)
+{
+ memcpy(data, zone, sizeof(struct blk_zone));
+ return 0;
+}
+
+static int blkdev_report_zone_fallback(struct block_device *bdev,
+ sector_t sector, struct blk_zone *zone)
{
- struct disk_report_zones_cb_args args = {
- .disk = disk,
+ struct blk_report_zones_args args = {
+ .cb = blkdev_report_zone_cb,
+ .data = zone,
+ .report_active = true,
};
- return disk->fops->report_zones(disk, sector, 1,
- disk_report_zones_cb, &args);
+ return blkdev_do_report_zones(bdev, sector, 1, &args);
}
-static bool blk_zone_wplug_handle_reset_or_finish(struct bio *bio,
- unsigned int wp_offset)
+/**
+ * blkdev_get_zone_info - Get a single zone information from cached data
+ * @bdev: Target block device
+ * @sector: Sector contained by the target zone
+ * @zone: zone structure to return the zone information
+ *
+ * Description:
+ * Get the zone information for the zone containing @sector using the zone
+ * write plug of the target zone, if one exist, or the disk zone condition
+ * array otherwise. The zone condition may be reported as being
+ * the BLK_ZONE_COND_ACTIVE condition for a zone that is in the implicit
+ * open, explicit open or closed condition.
+ *
+ * Returns 0 on success and a negative error code on failure.
+ */
+int blkdev_get_zone_info(struct block_device *bdev, sector_t sector,
+ struct blk_zone *zone)
{
- struct gendisk *disk = bio->bi_bdev->bd_disk;
- sector_t sector = bio->bi_iter.bi_sector;
+ struct gendisk *disk = bdev->bd_disk;
+ sector_t zone_sectors = bdev_zone_sectors(bdev);
struct blk_zone_wplug *zwplug;
unsigned long flags;
+ u8 *zones_cond;
- /* Conventional zones cannot be reset nor finished. */
- if (!bdev_zone_is_seq(bio->bi_bdev, sector)) {
- bio_io_error(bio);
- return true;
+ if (!bdev_is_zoned(bdev))
+ return -EOPNOTSUPP;
+
+ if (sector >= get_capacity(disk))
+ return -EINVAL;
+
+ memset(zone, 0, sizeof(*zone));
+ sector = ALIGN_DOWN(sector, zone_sectors);
+
+ rcu_read_lock();
+ zones_cond = rcu_dereference(disk->zones_cond);
+ if (!disk->zone_wplugs_hash || !zones_cond) {
+ rcu_read_unlock();
+ return blkdev_report_zone_fallback(bdev, sector, zone);
}
+ zone->cond = zones_cond[disk_zone_no(disk, sector)];
+ rcu_read_unlock();
+
+ zone->start = sector;
+ zone->len = zone_sectors;
/*
- * No-wait reset or finish BIOs do not make much sense as the callers
- * issue these as blocking operations in most cases. To avoid issues
- * the BIO execution potentially failing with BLK_STS_AGAIN, warn about
- * REQ_NOWAIT being set and ignore that flag.
+ * If this is a conventional zone, we do not have a zone write plug and
+ * can report the zone immediately.
*/
- if (WARN_ON_ONCE(bio->bi_opf & REQ_NOWAIT))
- bio->bi_opf &= ~REQ_NOWAIT;
+ if (zone->cond == BLK_ZONE_COND_NOT_WP) {
+ zone->type = BLK_ZONE_TYPE_CONVENTIONAL;
+ zone->capacity = zone_sectors;
+ zone->wp = ULLONG_MAX;
+ return 0;
+ }
+
+ /*
+ * This is a sequential write required zone. If the zone is read-only or
+ * offline, only set the zone write pointer to an invalid value and
+ * report the zone.
+ */
+ zone->type = BLK_ZONE_TYPE_SEQWRITE_REQ;
+ if (disk_zone_is_last(disk, zone))
+ zone->capacity = disk->last_zone_capacity;
+ else
+ zone->capacity = disk->zone_capacity;
+
+ if (zone->cond == BLK_ZONE_COND_READONLY ||
+ zone->cond == BLK_ZONE_COND_OFFLINE) {
+ zone->wp = ULLONG_MAX;
+ return 0;
+ }
/*
- * If we have a zone write plug, set its write pointer offset to 0
- * (reset case) or to the zone size (finish case). This will abort all
- * BIOs plugged for the target zone. It is fine as resetting or
- * finishing zones while writes are still in-flight will result in the
+ * If the zone does not have a zone write plug, it is either full or
+ * empty, as we otherwise would have a zone write plug for it. In this
+ * case, set the write pointer accordingly and report the zone.
+ * Otherwise, if we have a zone write plug, use it.
+ */
+ zwplug = disk_get_zone_wplug(disk, sector);
+ if (!zwplug) {
+ if (zone->cond == BLK_ZONE_COND_FULL)
+ zone->wp = ULLONG_MAX;
+ else
+ zone->wp = sector;
+ return 0;
+ }
+
+ spin_lock_irqsave(&zwplug->lock, flags);
+ if (zwplug->flags & BLK_ZONE_WPLUG_NEED_WP_UPDATE) {
+ spin_unlock_irqrestore(&zwplug->lock, flags);
+ disk_put_zone_wplug(zwplug);
+ return blkdev_report_zone_fallback(bdev, sector, zone);
+ }
+ zone->cond = zwplug->cond;
+ zone->wp = sector + zwplug->wp_offset;
+ spin_unlock_irqrestore(&zwplug->lock, flags);
+
+ disk_put_zone_wplug(zwplug);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(blkdev_get_zone_info);
+
+/**
+ * blkdev_report_zones_cached - Get cached zones information
+ * @bdev: Target block device
+ * @sector: Sector from which to report zones
+ * @nr_zones: Maximum number of zones to report
+ * @cb: Callback function called for each reported zone
+ * @data: Private data for the callback function
+ *
+ * Description:
+ * Similar to blkdev_report_zones() but instead of calling into the low level
+ * device driver to get the zone report from the device, use
+ * blkdev_get_zone_info() to generate the report from the disk zone write
+ * plugs and zones condition array. Since calling this function without a
+ * callback does not make sense, @cb must be specified.
+ */
+int blkdev_report_zones_cached(struct block_device *bdev, sector_t sector,
+ unsigned int nr_zones, report_zones_cb cb, void *data)
+{
+ struct gendisk *disk = bdev->bd_disk;
+ sector_t capacity = get_capacity(disk);
+ sector_t zone_sectors = bdev_zone_sectors(bdev);
+ unsigned int idx = 0;
+ struct blk_zone zone;
+ int ret;
+
+ if (!cb || !bdev_is_zoned(bdev) ||
+ WARN_ON_ONCE(!disk->fops->report_zones))
+ return -EOPNOTSUPP;
+
+ if (!nr_zones || sector >= capacity)
+ return 0;
+
+ /*
+ * If we do not have any zone write plug resources, fallback to using
+ * the regular zone report.
+ */
+ if (!disk_need_zone_resources(disk)) {
+ struct blk_report_zones_args args = {
+ .cb = cb,
+ .data = data,
+ .report_active = true,
+ };
+
+ return blkdev_do_report_zones(bdev, sector, nr_zones, &args);
+ }
+
+ for (sector = ALIGN_DOWN(sector, zone_sectors);
+ sector < capacity && idx < nr_zones;
+ sector += zone_sectors, idx++) {
+ ret = blkdev_get_zone_info(bdev, sector, &zone);
+ if (ret)
+ return ret;
+
+ ret = cb(&zone, idx, data);
+ if (ret)
+ return ret;
+ }
+
+ return idx;
+}
+EXPORT_SYMBOL_GPL(blkdev_report_zones_cached);
+
+static void blk_zone_reset_bio_endio(struct bio *bio)
+{
+ struct gendisk *disk = bio->bi_bdev->bd_disk;
+ sector_t sector = bio->bi_iter.bi_sector;
+ struct blk_zone_wplug *zwplug;
+
+ /*
+ * If we have a zone write plug, set its write pointer offset to 0.
+ * This will abort all BIOs plugged for the target zone. It is fine as
+ * resetting zones while writes are still in-flight will result in the
* writes failing anyway.
*/
zwplug = disk_get_zone_wplug(disk, sector);
if (zwplug) {
+ unsigned long flags;
+
spin_lock_irqsave(&zwplug->lock, flags);
- disk_zone_wplug_set_wp_offset(disk, zwplug, wp_offset);
+ disk_zone_wplug_set_wp_offset(disk, zwplug, 0);
spin_unlock_irqrestore(&zwplug->lock, flags);
disk_put_zone_wplug(zwplug);
+ } else {
+ disk_zone_set_cond(disk, sector, BLK_ZONE_COND_EMPTY);
}
-
- return false;
}
-static bool blk_zone_wplug_handle_reset_all(struct bio *bio)
+static void blk_zone_reset_all_bio_endio(struct bio *bio)
{
struct gendisk *disk = bio->bi_bdev->bd_disk;
+ sector_t capacity = get_capacity(disk);
struct blk_zone_wplug *zwplug;
unsigned long flags;
sector_t sector;
+ unsigned int i;
- /*
- * Set the write pointer offset of all zone write plugs to 0. This will
- * abort all plugged BIOs. It is fine as resetting zones while writes
- * are still in-flight will result in the writes failing anyway.
- */
- for (sector = 0; sector < get_capacity(disk);
- sector += disk->queue->limits.chunk_sectors) {
- zwplug = disk_get_zone_wplug(disk, sector);
- if (zwplug) {
+ /* Update the condition of all zone write plugs. */
+ rcu_read_lock();
+ for (i = 0; i < disk_zone_wplugs_hash_size(disk); i++) {
+ hlist_for_each_entry_rcu(zwplug, &disk->zone_wplugs_hash[i],
+ node) {
spin_lock_irqsave(&zwplug->lock, flags);
disk_zone_wplug_set_wp_offset(disk, zwplug, 0);
spin_unlock_irqrestore(&zwplug->lock, flags);
- disk_put_zone_wplug(zwplug);
}
}
+ rcu_read_unlock();
- return false;
+ /* Update the cached zone conditions. */
+ for (sector = 0; sector < capacity;
+ sector += bdev_zone_sectors(bio->bi_bdev))
+ disk_zone_set_cond(disk, sector, BLK_ZONE_COND_EMPTY);
+}
+
+static void blk_zone_finish_bio_endio(struct bio *bio)
+{
+ struct block_device *bdev = bio->bi_bdev;
+ struct gendisk *disk = bdev->bd_disk;
+ sector_t sector = bio->bi_iter.bi_sector;
+ struct blk_zone_wplug *zwplug;
+
+ /*
+ * If we have a zone write plug, set its write pointer offset to the
+ * zone size. This will abort all BIOs plugged for the target zone. It
+ * is fine as resetting zones while writes are still in-flight will
+ * result in the writes failing anyway.
+ */
+ zwplug = disk_get_zone_wplug(disk, sector);
+ if (zwplug) {
+ unsigned long flags;
+
+ spin_lock_irqsave(&zwplug->lock, flags);
+ disk_zone_wplug_set_wp_offset(disk, zwplug,
+ bdev_zone_sectors(bdev));
+ spin_unlock_irqrestore(&zwplug->lock, flags);
+ disk_put_zone_wplug(zwplug);
+ } else {
+ disk_zone_set_cond(disk, sector, BLK_ZONE_COND_FULL);
+ }
+}
+
+void blk_zone_mgmt_bio_endio(struct bio *bio)
+{
+ /* If the BIO failed, we have nothing to do. */
+ if (bio->bi_status != BLK_STS_OK)
+ return;
+
+ switch (bio_op(bio)) {
+ case REQ_OP_ZONE_RESET:
+ blk_zone_reset_bio_endio(bio);
+ return;
+ case REQ_OP_ZONE_RESET_ALL:
+ blk_zone_reset_all_bio_endio(bio);
+ return;
+ case REQ_OP_ZONE_FINISH:
+ blk_zone_finish_bio_endio(bio);
+ return;
+ default:
+ return;
+ }
}
static void disk_zone_wplug_schedule_bio_work(struct gendisk *disk,
@@ -836,6 +1236,7 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
*/
void blk_zone_write_plug_bio_merged(struct bio *bio)
{
+ struct gendisk *disk = bio->bi_bdev->bd_disk;
struct blk_zone_wplug *zwplug;
unsigned long flags;
@@ -857,13 +1258,13 @@ void blk_zone_write_plug_bio_merged(struct bio *bio)
* have at least one request and one BIO referencing the zone write
* plug. So this should not fail.
*/
- zwplug = disk_get_zone_wplug(bio->bi_bdev->bd_disk,
- bio->bi_iter.bi_sector);
+ zwplug = disk_get_zone_wplug(disk, bio->bi_iter.bi_sector);
if (WARN_ON_ONCE(!zwplug))
return;
spin_lock_irqsave(&zwplug->lock, flags);
zwplug->wp_offset += bio_sectors(bio);
+ disk_zone_wplug_update_cond(disk, zwplug);
spin_unlock_irqrestore(&zwplug->lock, flags);
}
@@ -922,6 +1323,7 @@ void blk_zone_write_plug_init_request(struct request *req)
/* Drop the reference taken by disk_zone_wplug_add_bio(). */
blk_queue_exit(q);
zwplug->wp_offset += bio_sectors(bio);
+ disk_zone_wplug_update_cond(disk, zwplug);
req_back_sector += bio_sectors(bio);
}
@@ -985,6 +1387,7 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
/* Advance the zone write pointer offset. */
zwplug->wp_offset += bio_sectors(bio);
+ disk_zone_wplug_update_cond(disk, zwplug);
return true;
}
@@ -1106,6 +1509,30 @@ static void blk_zone_wplug_handle_native_zone_append(struct bio *bio)
disk_put_zone_wplug(zwplug);
}
+static bool blk_zone_wplug_handle_zone_mgmt(struct bio *bio)
+{
+ if (bio_op(bio) != REQ_OP_ZONE_RESET_ALL &&
+ !bdev_zone_is_seq(bio->bi_bdev, bio->bi_iter.bi_sector)) {
+ /*
+ * Zone reset and zone finish operations do not apply to
+ * conventional zones.
+ */
+ bio_io_error(bio);
+ return true;
+ }
+
+ /*
+ * No-wait zone management BIOs do not make much sense as the callers
+ * issue these as blocking operations in most cases. To avoid issues
+ * with the BIO execution potentially failing with BLK_STS_AGAIN, warn
+ * about REQ_NOWAIT being set and ignore that flag.
+ */
+ if (WARN_ON_ONCE(bio->bi_opf & REQ_NOWAIT))
+ bio->bi_opf &= ~REQ_NOWAIT;
+
+ return false;
+}
+
/**
* blk_zone_plug_bio - Handle a zone write BIO with zone write plugging
* @bio: The BIO being submitted
@@ -1153,12 +1580,9 @@ bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
case REQ_OP_WRITE_ZEROES:
return blk_zone_wplug_handle_write(bio, nr_segs);
case REQ_OP_ZONE_RESET:
- return blk_zone_wplug_handle_reset_or_finish(bio, 0);
case REQ_OP_ZONE_FINISH:
- return blk_zone_wplug_handle_reset_or_finish(bio,
- bdev_zone_sectors(bdev));
case REQ_OP_ZONE_RESET_ALL:
- return blk_zone_wplug_handle_reset_all(bio);
+ return blk_zone_wplug_handle_zone_mgmt(bio);
default:
return false;
}
@@ -1332,11 +1756,6 @@ put_zwplug:
disk_put_zone_wplug(zwplug);
}
-static inline unsigned int disk_zone_wplugs_hash_size(struct gendisk *disk)
-{
- return 1U << disk->zone_wplugs_hash_bits;
-}
-
void disk_init_zone_resources(struct gendisk *disk)
{
spin_lock_init(&disk->zone_wplugs_lock);
@@ -1417,22 +1836,16 @@ static void disk_destroy_zone_wplugs_hash_table(struct gendisk *disk)
disk->zone_wplugs_hash_bits = 0;
}
-static unsigned int disk_set_conv_zones_bitmap(struct gendisk *disk,
- unsigned long *bitmap)
+static void disk_set_zones_cond_array(struct gendisk *disk, u8 *zones_cond)
{
- unsigned int nr_conv_zones = 0;
unsigned long flags;
spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
- if (bitmap)
- nr_conv_zones = bitmap_weight(bitmap, disk->nr_zones);
- bitmap = rcu_replace_pointer(disk->conv_zones_bitmap, bitmap,
- lockdep_is_held(&disk->zone_wplugs_lock));
+ zones_cond = rcu_replace_pointer(disk->zones_cond, zones_cond,
+ lockdep_is_held(&disk->zone_wplugs_lock));
spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
- kfree_rcu_mightsleep(bitmap);
-
- return nr_conv_zones;
+ kfree_rcu_mightsleep(zones_cond);
}
void disk_free_zone_resources(struct gendisk *disk)
@@ -1456,31 +1869,37 @@ void disk_free_zone_resources(struct gendisk *disk)
mempool_destroy(disk->zone_wplugs_pool);
disk->zone_wplugs_pool = NULL;
- disk_set_conv_zones_bitmap(disk, NULL);
+ disk_set_zones_cond_array(disk, NULL);
disk->zone_capacity = 0;
disk->last_zone_capacity = 0;
disk->nr_zones = 0;
}
-static inline bool disk_need_zone_resources(struct gendisk *disk)
-{
- /*
- * All mq zoned devices need zone resources so that the block layer
- * can automatically handle write BIO plugging. BIO-based device drivers
- * (e.g. DM devices) are normally responsible for handling zone write
- * ordering and do not need zone resources, unless the driver requires
- * zone append emulation.
- */
- return queue_is_mq(disk->queue) ||
- queue_emulates_zone_append(disk->queue);
-}
+struct blk_revalidate_zone_args {
+ struct gendisk *disk;
+ u8 *zones_cond;
+ unsigned int nr_zones;
+ unsigned int nr_conv_zones;
+ unsigned int zone_capacity;
+ unsigned int last_zone_capacity;
+ sector_t sector;
+};
static int disk_revalidate_zone_resources(struct gendisk *disk,
- unsigned int nr_zones)
+ struct blk_revalidate_zone_args *args)
{
struct queue_limits *lim = &disk->queue->limits;
unsigned int pool_size;
+ args->disk = disk;
+ args->nr_zones =
+ DIV_ROUND_UP_ULL(get_capacity(disk), lim->chunk_sectors);
+
+ /* Cached zone conditions: 1 byte per zone */
+ args->zones_cond = kzalloc(args->nr_zones, GFP_NOIO);
+ if (!args->zones_cond)
+ return -ENOMEM;
+
if (!disk_need_zone_resources(disk))
return 0;
@@ -1490,7 +1909,8 @@ static int disk_revalidate_zone_resources(struct gendisk *disk,
*/
pool_size = max(lim->max_open_zones, lim->max_active_zones);
if (!pool_size)
- pool_size = min(BLK_ZONE_WPLUG_DEFAULT_POOL_SIZE, nr_zones);
+ pool_size =
+ min(BLK_ZONE_WPLUG_DEFAULT_POOL_SIZE, args->nr_zones);
if (!disk->zone_wplugs_hash)
return disk_alloc_zone_resources(disk, pool_size);
@@ -1498,15 +1918,6 @@ static int disk_revalidate_zone_resources(struct gendisk *disk,
return 0;
}
-struct blk_revalidate_zone_args {
- struct gendisk *disk;
- unsigned long *conv_zones_bitmap;
- unsigned int nr_zones;
- unsigned int zone_capacity;
- unsigned int last_zone_capacity;
- sector_t sector;
-};
-
/*
* Update the disk zone resources information and device queue limits.
* The disk queue is frozen when this is executed.
@@ -1515,30 +1926,34 @@ static int disk_update_zone_resources(struct gendisk *disk,
struct blk_revalidate_zone_args *args)
{
struct request_queue *q = disk->queue;
- unsigned int nr_seq_zones, nr_conv_zones;
- unsigned int pool_size;
+ unsigned int nr_seq_zones;
+ unsigned int pool_size, memflags;
struct queue_limits lim;
+ int ret = 0;
+
+ lim = queue_limits_start_update(q);
+
+ memflags = blk_mq_freeze_queue(q);
disk->nr_zones = args->nr_zones;
- disk->zone_capacity = args->zone_capacity;
- disk->last_zone_capacity = args->last_zone_capacity;
- nr_conv_zones =
- disk_set_conv_zones_bitmap(disk, args->conv_zones_bitmap);
- if (nr_conv_zones >= disk->nr_zones) {
+ if (args->nr_conv_zones >= disk->nr_zones) {
pr_warn("%s: Invalid number of conventional zones %u / %u\n",
- disk->disk_name, nr_conv_zones, disk->nr_zones);
- return -ENODEV;
+ disk->disk_name, args->nr_conv_zones, disk->nr_zones);
+ ret = -ENODEV;
+ goto unfreeze;
}
- lim = queue_limits_start_update(q);
+ disk->zone_capacity = args->zone_capacity;
+ disk->last_zone_capacity = args->last_zone_capacity;
+ disk_set_zones_cond_array(disk, args->zones_cond);
/*
- * Some devices can advertize zone resource limits that are larger than
+ * Some devices can advertise zone resource limits that are larger than
* the number of sequential zones of the zoned block device, e.g. a
* small ZNS namespace. For such case, assume that the zoned device has
* no zone resource limits.
*/
- nr_seq_zones = disk->nr_zones - nr_conv_zones;
+ nr_seq_zones = disk->nr_zones - args->nr_conv_zones;
if (lim.max_open_zones >= nr_seq_zones)
lim.max_open_zones = 0;
if (lim.max_active_zones >= nr_seq_zones)
@@ -1568,7 +1983,53 @@ static int disk_update_zone_resources(struct gendisk *disk,
}
commit:
- return queue_limits_commit_update_frozen(q, &lim);
+ ret = queue_limits_commit_update(q, &lim);
+
+unfreeze:
+ if (ret)
+ disk_free_zone_resources(disk);
+
+ blk_mq_unfreeze_queue(q, memflags);
+
+ return ret;
+}
+
+static int blk_revalidate_zone_cond(struct blk_zone *zone, unsigned int idx,
+ struct blk_revalidate_zone_args *args)
+{
+ enum blk_zone_cond cond = zone->cond;
+
+ /* Check that the zone condition is consistent with the zone type. */
+ switch (cond) {
+ case BLK_ZONE_COND_NOT_WP:
+ if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL)
+ goto invalid_condition;
+ break;
+ case BLK_ZONE_COND_IMP_OPEN:
+ case BLK_ZONE_COND_EXP_OPEN:
+ case BLK_ZONE_COND_CLOSED:
+ case BLK_ZONE_COND_EMPTY:
+ case BLK_ZONE_COND_FULL:
+ case BLK_ZONE_COND_OFFLINE:
+ case BLK_ZONE_COND_READONLY:
+ if (zone->type != BLK_ZONE_TYPE_SEQWRITE_REQ)
+ goto invalid_condition;
+ break;
+ default:
+ pr_warn("%s: Invalid zone condition 0x%X\n",
+ args->disk->disk_name, cond);
+ return -ENODEV;
+ }
+
+ blk_zone_set_cond(args->zones_cond, idx, cond);
+
+ return 0;
+
+invalid_condition:
+ pr_warn("%s: Invalid zone condition 0x%x for type 0x%x\n",
+ args->disk->disk_name, cond, zone->type);
+
+ return -ENODEV;
}
static int blk_revalidate_conv_zone(struct blk_zone *zone, unsigned int idx,
@@ -1585,17 +2046,7 @@ static int blk_revalidate_conv_zone(struct blk_zone *zone, unsigned int idx,
if (disk_zone_is_last(disk, zone))
args->last_zone_capacity = zone->capacity;
- if (!disk_need_zone_resources(disk))
- return 0;
-
- if (!args->conv_zones_bitmap) {
- args->conv_zones_bitmap =
- bitmap_zalloc(args->nr_zones, GFP_NOIO);
- if (!args->conv_zones_bitmap)
- return -ENOMEM;
- }
-
- set_bit(idx, args->conv_zones_bitmap);
+ args->nr_conv_zones++;
return 0;
}
@@ -1693,6 +2144,11 @@ static int blk_revalidate_zone_cb(struct blk_zone *zone, unsigned int idx,
return -ENODEV;
}
+ /* Check zone condition */
+ ret = blk_revalidate_zone_cond(zone, idx, args);
+ if (ret)
+ return ret;
+
/* Check zone type */
switch (zone->type) {
case BLK_ZONE_TYPE_CONVENTIONAL:
@@ -1733,7 +2189,11 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
sector_t zone_sectors = q->limits.chunk_sectors;
sector_t capacity = get_capacity(disk);
struct blk_revalidate_zone_args args = { };
- unsigned int noio_flag;
+ unsigned int memflags, noio_flag;
+ struct blk_report_zones_args rep_args = {
+ .cb = blk_revalidate_zone_cb,
+ .data = &args,
+ };
int ret = -ENOMEM;
if (WARN_ON_ONCE(!blk_queue_is_zoned(q)))
@@ -1756,17 +2216,14 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
* Ensure that all memory allocations in this context are done as if
* GFP_NOIO was specified.
*/
- args.disk = disk;
- args.nr_zones = (capacity + zone_sectors - 1) >> ilog2(zone_sectors);
noio_flag = memalloc_noio_save();
- ret = disk_revalidate_zone_resources(disk, args.nr_zones);
+ ret = disk_revalidate_zone_resources(disk, &args);
if (ret) {
memalloc_noio_restore(noio_flag);
return ret;
}
- ret = disk->fops->report_zones(disk, 0, UINT_MAX,
- blk_revalidate_zone_cb, &args);
+ ret = disk->fops->report_zones(disk, 0, UINT_MAX, &rep_args);
if (!ret) {
pr_warn("%s: No zones reported\n", disk->disk_name);
ret = -ENODEV;
@@ -1783,20 +2240,14 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
ret = -ENODEV;
}
- /*
- * Set the new disk zone parameters only once the queue is frozen and
- * all I/Os are completed.
- */
if (ret > 0)
- ret = disk_update_zone_resources(disk, &args);
- else
- pr_warn("%s: failed to revalidate zones\n", disk->disk_name);
- if (ret) {
- unsigned int memflags = blk_mq_freeze_queue(q);
+ return disk_update_zone_resources(disk, &args);
- disk_free_zone_resources(disk);
- blk_mq_unfreeze_queue(q, memflags);
- }
+ pr_warn("%s: failed to revalidate zones\n", disk->disk_name);
+
+ memflags = blk_mq_freeze_queue(q);
+ disk_free_zone_resources(disk);
+ blk_mq_unfreeze_queue(q, memflags);
return ret;
}
@@ -1817,6 +2268,7 @@ EXPORT_SYMBOL_GPL(blk_revalidate_disk_zones);
int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
sector_t nr_sects, gfp_t gfp_mask)
{
+ struct gendisk *disk = bdev->bd_disk;
int ret;
if (WARN_ON_ONCE(!bdev_is_zoned(bdev)))
@@ -1832,7 +2284,7 @@ int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
* pointer. Undo this using a report zone to update the zone write
* pointer to the correct current value.
*/
- ret = disk_zone_sync_wp_offset(bdev->bd_disk, sector);
+ ret = disk->fops->report_zones(disk, sector, 1, NULL);
if (ret != 1)
return ret < 0 ? ret : -EIO;
@@ -1851,18 +2303,22 @@ static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
unsigned int zwp_wp_offset, zwp_flags;
unsigned int zwp_zone_no, zwp_ref;
unsigned int zwp_bio_list_size;
+ enum blk_zone_cond zwp_cond;
unsigned long flags;
spin_lock_irqsave(&zwplug->lock, flags);
zwp_zone_no = zwplug->zone_no;
zwp_flags = zwplug->flags;
zwp_ref = refcount_read(&zwplug->ref);
+ zwp_cond = zwplug->cond;
zwp_wp_offset = zwplug->wp_offset;
zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
spin_unlock_irqrestore(&zwplug->lock, flags);
- seq_printf(m, "%u 0x%x %u %u %u\n", zwp_zone_no, zwp_flags, zwp_ref,
- zwp_wp_offset, zwp_bio_list_size);
+ seq_printf(m,
+ "Zone no: %u, flags: 0x%x, ref: %u, cond: %s, wp ofst: %u, pending BIO: %u\n",
+ zwp_zone_no, zwp_flags, zwp_ref, blk_zone_cond_str(zwp_cond),
+ zwp_wp_offset, zwp_bio_list_size);
}
int queue_zone_wplugs_show(void *data, struct seq_file *m)
diff --git a/block/blk.h b/block/blk.h
index 32a10024efba..4d809588b771 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -489,10 +489,24 @@ 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.
*/
diff --git a/block/ioctl.c b/block/ioctl.c
index 3927ca4707d0..698629e4c619 100644
--- a/block/ioctl.c
+++ b/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:
diff --git a/drivers/block/null_blk/null_blk.h b/drivers/block/null_blk/null_blk.h
index 7bb6128dbaaf..6c4c4bbe7dad 100644
--- a/drivers/block/null_blk/null_blk.h
+++ b/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,
diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 4e5728f45989..6a93b12a06ff 100644
--- a/drivers/block/null_blk/zoned.c
+++ b/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;
}
diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 96e07763cd28..97cc4bc0a6ce 100644
--- a/drivers/block/ublk_drv.c
+++ b/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 @@ free_req:
if (!zone->len)
break;
- ret = cb(zone, i, data);
+ ret = disk_report_zone(disk, zone, i, args);
if (ret)
goto out;
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index f061420dfb10..a5e97f03dbf0 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -584,7 +584,8 @@ out:
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;
diff --git a/drivers/block/zloop.c b/drivers/block/zloop.c
index a423228e201b..92be9f0af00a 100644
--- a/drivers/block/zloop.c
+++ b/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;
}
diff --git a/drivers/md/dm-zone.c b/drivers/md/dm-zone.c
index 78e17dd4d01b..984fb621b0e9 100644
--- a/drivers/md/dm-zone.c
+++ b/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);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 245f52b59215..7a795979ec72 100644
--- a/drivers/md/dm.h
+++ b/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,
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index fa4181d7de73..c0fe50fb7b08 100644
--- a/drivers/nvme/host/core.c
+++ b/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
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 543e17aead12..0b7ac0735bd0 100644
--- a/drivers/nvme/host/multipath.c
+++ b/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;
}
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 102fae6a231c..928c748ccbd1 100644
--- a/drivers/nvme/host/nvme.h
+++ b/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,
diff --git a/drivers/nvme/host/zns.c b/drivers/nvme/host/zns.c
index cce4c5b55aa9..deea2dbef5b8 100644
--- a/drivers/nvme/host/zns.c
+++ b/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++;
diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index 36382eca941c..574af8243016 100644
--- a/drivers/scsi/sd.h
+++ b/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 */
diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c
index a8db66428f80..56e455fb5add 100644
--- a/drivers/scsi/sd_zbc.c
+++ b/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;
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index 0ea0df18a8e4..a16b1a896c78 100644
--- a/fs/btrfs/zoned.c
+++ b/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))
diff --git a/fs/xfs/libxfs/xfs_zones.c b/fs/xfs/libxfs/xfs_zones.c
index b0791a71931c..b40f71f878b5 100644
--- a/fs/xfs/libxfs/xfs_zones.c
+++ b/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);
diff --git a/fs/xfs/xfs_zone_alloc.c b/fs/xfs/xfs_zone_alloc.c
index 1147bacb2da8..d121768dbccb 100644
--- a/fs/xfs/xfs_zone_alloc.c
+++ b/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)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 99be263b31ab..f0ab02e0a673 100644
--- a/include/linux/blkdev.h
+++ b/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],
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 84fdc3a6a19a..38f625af6ab4 100644
--- a/include/linux/device-mapper.h
+++ b/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;
};
diff --git a/include/uapi/linux/blkzoned.h b/include/uapi/linux/blkzoned.h
index f85743ef6e7d..e33f02703350 100644
--- a/include/uapi/linux/blkzoned.h
+++ b/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 */
diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h
index 957ce3343a4f..66ca526cf786 100644
--- a/include/uapi/linux/fs.h
+++ b/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)