[PATCH] zloop: restore finished zones with reduced capacity

[PATCH] zloop: restore finished zones with reduced capacity

[raoxu]

From: Xu Rao <raoxu@uniontech.com>

zloop_finish_zone() truncates a finished sequential zone backing file to
zone_size. When zone_capacity is smaller than zone_size,
zloop_update_seq_zone() rejects that file when the device is re-added
because it treats every file larger than zone_capacity as invalid.

Documentation/admin-guide/blockdev/zoned_loop.rst, section "Overview",
states that a zone finish operation truncates the backing file to the zone
size. A file whose size equals zone_size is therefore a valid
representation of a finished zone.

Accept both zone_capacity and zone_size as full-zone representations. Keep
rejecting all other files larger than zone_capacity.

Fixes: eb0570c7df23 ("block: new zoned loop block device driver")
Cc: stable@vger.kernel.org
Signed-off-by: Xu Rao <raoxu@uniontech.com>
---
 drivers/block/zloop.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/block/zloop.c b/drivers/block/zloop.c
index 55eeb6aac0ea..b69d798f203c 100644
--- a/drivers/block/zloop.c
+++ b/drivers/block/zloop.c
@@ -324,7 +324,8 @@ static int zloop_update_seq_zone(struct zloop_device *zlo, unsigned int zone_no)
 	}
 
 	file_sectors = stat.size >> SECTOR_SHIFT;
-	if (file_sectors > zlo->zone_capacity) {
+	if (file_sectors > zlo->zone_capacity &&
+	    file_sectors != zlo->zone_size) {
 		pr_err("Zone %u file too large (%llu sectors > %llu)\n",
 		       zone_no, file_sectors, zlo->zone_capacity);
 		return -EINVAL;
@@ -339,7 +340,8 @@ static int zloop_update_seq_zone(struct zloop_device *zlo, unsigned int zone_no)
 	spin_lock(&zone->wp_lock);
 	if (!file_sectors) {
 		zloop_mark_empty(zlo, zone);
-	} else if (file_sectors == zlo->zone_capacity) {
+	} else if (file_sectors == zlo->zone_capacity ||
+		   file_sectors == zlo->zone_size) {
 		zloop_mark_full(zlo, zone);
 	} else {
 		if (zone->cond != BLK_ZONE_COND_IMP_OPEN &&
-- 
2.50.1

Re: [PATCH] zloop: restore finished zones with reduced capacity

[Damien Le Moal]

On 7/31/26 15:11, raoxu wrote:
> From: Xu Rao <raoxu@uniontech.com>
> 
> zloop_finish_zone() truncates a finished sequential zone backing file to
> zone_size. When zone_capacity is smaller than zone_size,
> zloop_update_seq_zone() rejects that file when the device is re-added
> because it treats every file larger than zone_capacity as invalid.
> 
> Documentation/admin-guide/blockdev/zoned_loop.rst, section "Overview",
> states that a zone finish operation truncates the backing file to the zone
> size. A file whose size equals zone_size is therefore a valid
> representation of a finished zone.
> 
> Accept both zone_capacity and zone_size as full-zone representations. Keep
> rejecting all other files larger than zone_capacity.
> 
> Fixes: eb0570c7df23 ("block: new zoned loop block device driver")
> Cc: stable@vger.kernel.org
> Signed-off-by: Xu Rao <raoxu@uniontech.com>
> ---
>  drivers/block/zloop.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/block/zloop.c b/drivers/block/zloop.c
> index 55eeb6aac0ea..b69d798f203c 100644
> --- a/drivers/block/zloop.c
> +++ b/drivers/block/zloop.c
> @@ -324,7 +324,8 @@ static int zloop_update_seq_zone(struct zloop_device *zlo, unsigned int zone_no)
>  	}
>  
>  	file_sectors = stat.size >> SECTOR_SHIFT;
> -	if (file_sectors > zlo->zone_capacity) {
> +	if (file_sectors > zlo->zone_capacity &&
> +	    file_sectors != zlo->zone_size) {
>  		pr_err("Zone %u file too large (%llu sectors > %llu)\n",
>  		       zone_no, file_sectors, zlo->zone_capacity);
>  		return -EINVAL;
> @@ -339,7 +340,8 @@ static int zloop_update_seq_zone(struct zloop_device *zlo, unsigned int zone_no)
>  	spin_lock(&zone->wp_lock);
>  	if (!file_sectors) {
>  		zloop_mark_empty(zlo, zone);
> -	} else if (file_sectors == zlo->zone_capacity) {
> +	} else if (file_sectors == zlo->zone_capacity ||
> +		   file_sectors == zlo->zone_size) {
>  		zloop_mark_full(zlo, zone);
>  	} else {
>  		if (zone->cond != BLK_ZONE_COND_IMP_OPEN &&

The problem comes from at least zloop_finish_zone() which truncates a zone file
to zlo->zone_size << SECTOR_SHIFT instead of zlo->zone_capacity << SECTOR_SHIFT.
So I think that is what we need to fix.

-- 
Damien Le Moal
Western Digital Research

[PATCH v2] zloop: truncate finished zones to zone capacity

[raoxu]

From: Xu Rao <raoxu@uniontech.com>

The size of a sequential zone backing file records the amount of data
written and is used to restore the zone state. A backing file whose size
is equal to the zone capacity is restored as a full zone, while a file
larger than the zone capacity is rejected as invalid.

However, zloop_finish_zone() currently truncates the backing file to the
zone size. For devices with a reduced zone capacity, finishing a zone
therefore creates a backing file larger than the zone capacity. After the
device is removed and later re-added, that zone file is rejected instead
of being restored as a full zone.

Truncate finished sequential zones to the zone capacity, matching the
persistent representation accepted by zloop_update_seq_zone() for a full
zone.

Suggested-by: Damien Le Moal <dlemoal@kernel.org>
Fixes: eb0570c7df23 ("block: new zoned loop block device driver")
Cc: stable@vger.kernel.org
Signed-off-by: Xu Rao <raoxu@uniontech.com>
---
Changes in v2:
- Fix zloop_finish_zone() instead of accepting oversized backing files in
  zloop_update_seq_zone(), as suggested by Damien.
- Update documentation to describe the corrected backing file size.

 Documentation/admin-guide/blockdev/zoned_loop.rst | 9 ++++-----
 drivers/block/zloop.c                             | 3 ++-
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/Documentation/admin-guide/blockdev/zoned_loop.rst b/Documentation/admin-guide/blockdev/zoned_loop.rst
index f4f1f3121bf9..95974425ca05 100644
--- a/Documentation/admin-guide/blockdev/zoned_loop.rst
+++ b/Documentation/admin-guide/blockdev/zoned_loop.rst
@@ -30,11 +30,10 @@ indicates the position of the write pointer of the zone.

 When resetting a sequential zone, its backing file size is truncated to zero.
 Conversely, for a zone finish operation, the backing file is truncated to the
-zone size. With this, the maximum capacity of a zloop zoned block device created
-can be larger configured to be larger than the storage space available on the
-backing file system. Of course, for such configuration, writing more data than
-the storage space available on the backing file system will result in write
-errors.
+zone capacity. With this, a zloop zoned block device can be configured with a
+larger capacity than the storage space available on the backing file system. Of
+course, for such configuration, writing more data than the storage space
+available on the backing file system will result in write errors.

 The zoned loop block device driver implements a complete zone transition state
 machine. That is, zones can be empty, implicitly opened, explicitly opened,
diff --git a/drivers/block/zloop.c b/drivers/block/zloop.c
index 55eeb6aac0ea..1333fae05ac5 100644
--- a/drivers/block/zloop.c
+++ b/drivers/block/zloop.c
@@ -478,7 +478,8 @@ static int zloop_finish_zone(struct zloop_device *zlo, unsigned int zone_no)
 	    zone->cond == BLK_ZONE_COND_FULL)
 		goto unlock;

-	if (vfs_truncate(&zone->file->f_path, zlo->zone_size << SECTOR_SHIFT)) {
+	if (vfs_truncate(&zone->file->f_path,
+			 zlo->zone_capacity << SECTOR_SHIFT)) {
 		set_bit(ZLOOP_ZONE_SEQ_ERROR, &zone->flags);
 		ret = -EIO;
 		goto unlock;
--
2.50.1

Re: [PATCH v2] zloop: truncate finished zones to zone capacity

[Damien Le Moal]

On 8/4/26 11:34, raoxu wrote:
> From: Xu Rao <raoxu@uniontech.com>
> 
> The size of a sequential zone backing file records the amount of data
> written and is used to restore the zone state. A backing file whose size
> is equal to the zone capacity is restored as a full zone, while a file
> larger than the zone capacity is rejected as invalid.
> 
> However, zloop_finish_zone() currently truncates the backing file to the
> zone size. For devices with a reduced zone capacity, finishing a zone
> therefore creates a backing file larger than the zone capacity. After the
> device is removed and later re-added, that zone file is rejected instead
> of being restored as a full zone.
> 
> Truncate finished sequential zones to the zone capacity, matching the
> persistent representation accepted by zloop_update_seq_zone() for a full
> zone.
> 
> Suggested-by: Damien Le Moal <dlemoal@kernel.org>
> Fixes: eb0570c7df23 ("block: new zoned loop block device driver")
> Cc: stable@vger.kernel.org
> Signed-off-by: Xu Rao <raoxu@uniontech.com>

Looks good.

Reviewed-by: Damien Le Moal <dlemoal@kernel.org>

-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v2] zloop: truncate finished zones to zone capacity

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2] zloop: truncate finished zones to zone capacity

[Christoph Hellwig]

On Tue, Aug 04, 2026 at 10:34:03AM +0800, raoxu wrote:
> However, zloop_finish_zone() currently truncates the backing file to the
> zone size. For devices with a reduced zone capacity, finishing a zone
> therefore creates a backing file larger than the zone capacity. After the
> device is removed and later re-added, that zone file is rejected instead
> of being restored as a full zone.

Can you add a blktests test case for this?

Re: [PATCH v2] zloop: truncate finished zones to zone capacity

[Jens Axboe]

On Tue, 04 Aug 2026 10:34:03 +0800, raoxu wrote:
> The size of a sequential zone backing file records the amount of data
> written and is used to restore the zone state. A backing file whose size
> is equal to the zone capacity is restored as a full zone, while a file
> larger than the zone capacity is rejected as invalid.
> 
> However, zloop_finish_zone() currently truncates the backing file to the
> zone size. For devices with a reduced zone capacity, finishing a zone
> therefore creates a backing file larger than the zone capacity. After the
> device is removed and later re-added, that zone file is rejected instead
> of being restored as a full zone.
> 
> [...]

Applied, thanks!

[1/1] zloop: truncate finished zones to zone capacity
      commit: 72e67c118642634c25465db0c8bcfa54c4ce086c

Best regards,
-- 
Jens Axboe