block: rework zone reporting

Avoid the need to allocate a potentially large array of struct blk_zone
in the block layer by switching the ->report_zones method interface to
a callback model. Now the caller simply supplies a callback that is
executed on each reported zone, and private data for it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

1 files changed, 97 insertions, 156 deletions

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index ae665e490858..6fad6f3f6980 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -101,44 +101,35 @@ EXPORT_SYMBOL_GPL(blkdev_nr_zones);
  * blkdev_report_zones - Get zones information
  * @bdev:	Target block device
  * @sector:	Sector from which to report zones
- * @zones:	Array of zone structures where to return the zones information
- * @nr_zones:	Number of zone structures in the zone array
+ * @nr_zones:	Maximum number of zones to report
+ * @cb:		Callback function called for each reported zone
+ * @data:	Private data for the callback
  *
  * Description:
- *    Get zone information starting from the zone containing @sector.
- *    The number of zone information reported may be less than the number
- *    requested by @nr_zones. The number of zones actually reported is
- *    returned in @nr_zones.
- *    The caller must use memalloc_noXX_save/restore() calls to control
- *    memory allocations done within this function (zone array and command
- *    buffer allocation by the device driver).
+ *    Get zone information starting from the zone containing @sector for at most
+ *    @nr_zones, and call @cb for each zone reported by the device.
+ *    To report all zones in a device starting from @sector, the BLK_ALL_ZONES
+ *    constant can be passed to @nr_zones.
+ *    Returns the number of zones reported by the device, or a negative errno
+ *    value in case of failure.
+ *
+ *    Note: The caller must use memalloc_noXX_save/restore() calls to control
+ *    memory allocations done within this function.
  */
 int blkdev_report_zones(struct block_device *bdev, sector_t sector,
-			struct blk_zone *zones, unsigned int *nr_zones)
+			unsigned int nr_zones, report_zones_cb cb, void *data)
 {
-	struct request_queue *q = bdev_get_queue(bdev);
 	struct gendisk *disk = bdev->bd_disk;
 	sector_t capacity = get_capacity(disk);
 
-	if (!blk_queue_is_zoned(q))
-		return -EOPNOTSUPP;
-
-	/*
-	 * A block device that advertized itself as zoned must have a
-	 * report_zones method. If it does not have one defined, the device
-	 * driver has a bug. So warn about that.
-	 */
-	if (WARN_ON_ONCE(!disk->fops->report_zones))
+	if (!blk_queue_is_zoned(bdev_get_queue(bdev)) ||
+	    WARN_ON_ONCE(!disk->fops->report_zones))
 		return -EOPNOTSUPP;
 
-	if (!*nr_zones || sector >= capacity) {
-		*nr_zones = 0;
+	if (!nr_zones || sector >= capacity)
 		return 0;
-	}
 
-	*nr_zones = min(*nr_zones, __blkdev_nr_zones(q, capacity - sector));
-
-	return disk->fops->report_zones(disk, sector, zones, nr_zones);
+	return disk->fops->report_zones(disk, sector, nr_zones, cb, data);
 }
 EXPORT_SYMBOL_GPL(blkdev_report_zones);
 
@@ -232,6 +223,20 @@ int blkdev_zone_mgmt(struct block_device *bdev, enum req_opf op,
 }
 EXPORT_SYMBOL_GPL(blkdev_zone_mgmt);
 
+struct zone_report_args {
+	struct blk_zone __user *zones;
+};
+
+static int blkdev_copy_zone_to_user(struct blk_zone *zone, unsigned int idx,
+				    void *data)
+{
+	struct zone_report_args *args = data;
+
+	if (copy_to_user(&args->zones[idx], zone, sizeof(struct blk_zone)))
+		return -EFAULT;
+	return 0;
+}
+
 /*
  * BLKREPORTZONE ioctl processing.
  * Called from blkdev_ioctl.
@@ -240,9 +245,9 @@ int blkdev_report_zones_ioctl(struct block_device *bdev, fmode_t mode,
 			      unsigned int cmd, unsigned long arg)
 {
 	void __user *argp = (void __user *)arg;
+	struct zone_report_args args;
 	struct request_queue *q;
 	struct blk_zone_report rep;
-	struct blk_zone *zones;
 	int ret;
 
 	if (!argp)
@@ -264,32 +269,16 @@ int blkdev_report_zones_ioctl(struct block_device *bdev, fmode_t mode,
 	if (!rep.nr_zones)
 		return -EINVAL;
 
-	rep.nr_zones = min(blkdev_nr_zones(bdev), rep.nr_zones);
-
-	zones = kvmalloc_array(rep.nr_zones, sizeof(struct blk_zone),
-			       GFP_KERNEL | __GFP_ZERO);
-	if (!zones)
-		return -ENOMEM;
-
-	ret = blkdev_report_zones(bdev, rep.sector, zones, &rep.nr_zones);
-	if (ret)
-		goto out;
-
-	if (copy_to_user(argp, &rep, sizeof(struct blk_zone_report))) {
-		ret = -EFAULT;
-		goto out;
-	}
-
-	if (rep.nr_zones) {
-		if (copy_to_user(argp + sizeof(struct blk_zone_report), zones,
-				 sizeof(struct blk_zone) * rep.nr_zones))
-			ret = -EFAULT;
-	}
+	args.zones = argp + sizeof(struct blk_zone_report);
+	ret = blkdev_report_zones(bdev, rep.sector, rep.nr_zones,
+				  blkdev_copy_zone_to_user, &args);
+	if (ret < 0)
+		return ret;
 
- out:
-	kvfree(zones);
-
-	return ret;
+	rep.nr_zones = ret;
+	if (copy_to_user(argp, &rep, sizeof(struct blk_zone_report)))
+		return -EFAULT;
+	return 0;
 }
 
 /*
@@ -351,31 +340,6 @@ static inline unsigned long *blk_alloc_zone_bitmap(int node,
 			    GFP_NOIO, node);
 }
 
-/*
- * Allocate an array of struct blk_zone to get nr_zones zone information.
- * The allocated array may be smaller than nr_zones.
- */
-static struct blk_zone *blk_alloc_zones(unsigned int *nr_zones)
-{
-	struct blk_zone *zones;
-	size_t nrz = min(*nr_zones, BLK_ZONED_REPORT_MAX_ZONES);
-
-	/*
-	 * GFP_KERNEL here is meaningless as the caller task context has
-	 * the PF_MEMALLOC_NOIO flag set in blk_revalidate_disk_zones()
-	 * with memalloc_noio_save().
-	 */
-	zones = kvcalloc(nrz, sizeof(struct blk_zone), GFP_KERNEL);
-	if (!zones) {
-		*nr_zones = 0;
-		return NULL;
-	}
-
-	*nr_zones = nrz;
-
-	return zones;
-}
-
 void blk_queue_free_zone_bitmaps(struct request_queue *q)
 {
 	kfree(q->seq_zones_bitmap);
@@ -384,12 +348,21 @@ void blk_queue_free_zone_bitmaps(struct request_queue *q)
 	q->seq_zones_wlock = NULL;
 }
 
+struct blk_revalidate_zone_args {
+	struct gendisk	*disk;
+	unsigned long	*seq_zones_bitmap;
+	unsigned long	*seq_zones_wlock;
+	sector_t	sector;
+};
+
 /*
  * Helper function to check the validity of zones of a zoned block device.
  */
-static bool blk_zone_valid(struct gendisk *disk, struct blk_zone *zone,
-			   sector_t *sector)
+static int blk_revalidate_zone_cb(struct blk_zone *zone, unsigned int idx,
+				  void *data)
 {
+	struct blk_revalidate_zone_args *args = data;
+	struct gendisk *disk = args->disk;
 	struct request_queue *q = disk->queue;
 	sector_t zone_sectors = blk_queue_zone_sectors(q);
 	sector_t capacity = get_capacity(disk);
@@ -409,14 +382,14 @@ static bool blk_zone_valid(struct gendisk *disk, struct blk_zone *zone,
 	    zone->len > zone_sectors) {
 		pr_warn("%s: Invalid zoned device with larger last zone size\n",
 			disk->disk_name);
-		return false;
+		return -ENODEV;
 	}
 
 	/* Check for holes in the zone report */
-	if (zone->start != *sector) {
+	if (zone->start != args->sector) {
 		pr_warn("%s: Zone gap at sectors %llu..%llu\n",
-			disk->disk_name, *sector, zone->start);
-		return false;
+			disk->disk_name, args->sector, zone->start);
+		return -ENODEV;
 	}
 
 	/* Check zone type */
@@ -428,12 +401,38 @@ static bool blk_zone_valid(struct gendisk *disk, struct blk_zone *zone,
 	default:
 		pr_warn("%s: Invalid zone type 0x%x at sectors %llu\n",
 			disk->disk_name, (int)zone->type, zone->start);
-		return false;
+		return -ENODEV;
 	}
 
-	*sector += zone->len;
+	if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL)
+		set_bit(idx, args->seq_zones_bitmap);
 
-	return true;
+	args->sector += zone->len;
+	return 0;
+}
+
+static int blk_update_zone_info(struct gendisk *disk, unsigned int nr_zones,
+				struct blk_revalidate_zone_args *args)
+{
+	/*
+	 * Ensure that all memory allocations in this context are done as
+	 * if GFP_NOIO was specified.
+	 */
+	unsigned int noio_flag = memalloc_noio_save();
+	struct request_queue *q = disk->queue;
+	int ret;
+
+	args->seq_zones_wlock = blk_alloc_zone_bitmap(q->node, nr_zones);
+	if (!args->seq_zones_wlock)
+		return -ENOMEM;
+	args->seq_zones_bitmap = blk_alloc_zone_bitmap(q->node, nr_zones);
+	if (!args->seq_zones_bitmap)
+		return -ENOMEM;
+
+	ret = disk->fops->report_zones(disk, 0, nr_zones,
+				       blk_revalidate_zone_cb, args);
+	memalloc_noio_restore(noio_flag);
+	return ret;
 }
 
 /**
@@ -449,11 +448,7 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 {
 	struct request_queue *q = disk->queue;
 	unsigned int nr_zones = __blkdev_nr_zones(q, get_capacity(disk));
-	unsigned long *seq_zones_wlock = NULL, *seq_zones_bitmap = NULL;
-	unsigned int i, rep_nr_zones = 0, z = 0, nrz;
-	struct blk_zone *zones = NULL;
-	unsigned int noio_flag;
-	sector_t sector = 0;
+	struct blk_revalidate_zone_args args = { .disk = disk };
 	int ret = 0;
 
 	if (WARN_ON_ONCE(!blk_queue_is_zoned(q)))
@@ -468,82 +463,28 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 		return 0;
 	}
 
-	/*
-	 * Ensure that all memory allocations in this context are done as
-	 * if GFP_NOIO was specified.
-	 */
-	noio_flag = memalloc_noio_save();
-
-	if (!nr_zones)
-		goto update;
-
-	/* Allocate bitmaps */
-	ret = -ENOMEM;
-	seq_zones_wlock = blk_alloc_zone_bitmap(q->node, nr_zones);
-	if (!seq_zones_wlock)
-		goto out;
-	seq_zones_bitmap = blk_alloc_zone_bitmap(q->node, nr_zones);
-	if (!seq_zones_bitmap)
-		goto out;
-
-	/*
-	 * Get zone information to check the zones and initialize
-	 * seq_zones_bitmap.
-	 */
-	rep_nr_zones = nr_zones;
-	zones = blk_alloc_zones(&rep_nr_zones);
-	if (!zones)
-		goto out;
-
-	while (z < nr_zones) {
-		nrz = min(nr_zones - z, rep_nr_zones);
-		ret = disk->fops->report_zones(disk, sector, zones, &nrz);
-		if (ret)
-			goto out;
-		if (!nrz)
-			break;
-		for (i = 0; i < nrz; i++) {
-			if (!blk_zone_valid(disk, &zones[i], &sector)) {
-				ret = -ENODEV;
-				goto out;
-			}
-			if (zones[i].type != BLK_ZONE_TYPE_CONVENTIONAL)
-				set_bit(z, seq_zones_bitmap);
-			z++;
-		}
-	}
-
-	if (WARN_ON(z != nr_zones)) {
-		ret = -EIO;
-		goto out;
-	}
+	if (nr_zones)
+		ret = blk_update_zone_info(disk, nr_zones, &args);
 
-update:
 	/*
 	 * Install the new bitmaps, making sure the queue is stopped and
 	 * all I/Os are completed (i.e. a scheduler is not referencing the
 	 * bitmaps).
 	 */
 	blk_mq_freeze_queue(q);
-	q->nr_zones = nr_zones;
-	swap(q->seq_zones_wlock, seq_zones_wlock);
-	swap(q->seq_zones_bitmap, seq_zones_bitmap);
-	blk_mq_unfreeze_queue(q);
-
-out:
-	memalloc_noio_restore(noio_flag);
-
-	kvfree(zones);
-	kfree(seq_zones_wlock);
-	kfree(seq_zones_bitmap);
-
-	if (ret) {
+	if (ret >= 0) {
+		q->nr_zones = nr_zones;
+		swap(q->seq_zones_wlock, args.seq_zones_wlock);
+		swap(q->seq_zones_bitmap, args.seq_zones_bitmap);
+		ret = 0;
+	} else {
 		pr_warn("%s: failed to revalidate zones\n", disk->disk_name);
-		blk_mq_freeze_queue(q);
 		blk_queue_free_zone_bitmaps(q);
-		blk_mq_unfreeze_queue(q);
 	}
+	blk_mq_unfreeze_queue(q);
 
+	kfree(args.seq_zones_wlock);
+	kfree(args.seq_zones_bitmap);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(blk_revalidate_disk_zones);