[PATCH for-next] dm: split write BIOs on zone boundaries when zone append is not emulated

[PATCH for-next] dm: split write BIOs on zone boundaries when zone append is not emulated

[Shin'ichiro Kawasaki]

Since commit f70291411ba2 ("block: Introduce
bio_needs_zone_write_plugging()"), blk_mq_submit_bio() no longer splits
write BIOs to comply with the requirements of zoned block devices. This
change applies to write BIOs issued by zoned DM targets. Following this
modification, it is expected that DM targets themselves split write BIOs
for zoned block devices.

Commit 2df7168717b7 ("dm: Always split write BIOs to zoned device
limits") updates the device-mapper driver to perform splits for the
write BIOs. However, it did not address the cases where DM targets do
not emulate zone append, such as in the cases of dm-linear or dm-flakey.
For these targets, when the write BIOs span across zone boundaries, they
trigger WARN_ON_ONCE(bio_straddles_zones(bio)) in
blk_zone_wplug_handle_write(). This results in I/O errors. The errors
are reproduced by running blktests test case zbd/004 using zoned
dm-linear or dm-flakey devices.

To avoid the I/O errors, handle the write BIOs regardless whether DM
targets emulate zone append or not, so that all write BIOs are split at
zone boundaries. For that purpose, drop the check for zone append
emulation in dm_zone_bio_needs_split(). Its argument 'md' is no longer
used then drop it also.

Fixes: 2df7168717b7 ("dm: Always split write BIOs to zoned device limits")
Cc: stable@vger.kernel.org
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
---
 drivers/md/dm.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index ca889328fdfe..a64809f04241 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1785,8 +1785,7 @@ static void init_clone_info(struct clone_info *ci, struct dm_io *io,
 }
 
 #ifdef CONFIG_BLK_DEV_ZONED
-static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
-					   struct bio *bio)
+static inline bool dm_zone_bio_needs_split(struct bio *bio)
 {
 	/*
 	 * Special case the zone operations that cannot or should not be split.
@@ -1802,13 +1801,10 @@ static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
 	}
 
 	/*
-	 * Mapped devices that require zone append emulation will use the block
-	 * layer zone write plugging. In such case, we must split any large BIO
-	 * to the mapped device limits to avoid potential deadlocks with queue
-	 * freeze operations.
+	 * When mapped devices use the block layer zone write plugging, we must
+	 * split any large BIO to the mapped device limits to avoid potential
+	 * deadlocks with queue freeze operations.
 	 */
-	if (!dm_emulate_zone_append(md))
-		return false;
 	return bio_needs_zone_write_plugging(bio) || bio_straddles_zones(bio);
 }
 
@@ -1932,8 +1928,7 @@ static blk_status_t __send_zone_reset_all(struct clone_info *ci)
 }
 
 #else
-static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
-					   struct bio *bio)
+static inline bool dm_zone_bio_needs_split(struct bio *bio)
 {
 	return false;
 }
@@ -1960,7 +1955,7 @@ static void dm_split_and_process_bio(struct mapped_device *md,
 
 	is_abnormal = is_abnormal_io(bio);
 	if (static_branch_unlikely(&zoned_enabled)) {
-		need_split = is_abnormal || dm_zone_bio_needs_split(md, bio);
+		need_split = is_abnormal || dm_zone_bio_needs_split(bio);
 	} else {
 		need_split = is_abnormal;
 	}
-- 
2.50.1

Re: [PATCH for-next] dm: split write BIOs on zone boundaries when zone append is not emulated

[Damien Le Moal]

On 2025/07/17 11:32, Shin'ichiro Kawasaki wrote:
> Since commit f70291411ba2 ("block: Introduce
> bio_needs_zone_write_plugging()"), blk_mq_submit_bio() no longer splits
> write BIOs to comply with the requirements of zoned block devices. This
> change applies to write BIOs issued by zoned DM targets. Following this
> modification, it is expected that DM targets themselves split write BIOs
> for zoned block devices.

This description is confusing: if the underlying device of a DM target is a
blk-mq block device, then blk_mq_submit_bio() will be called and will split BIOs
as needed, regardless of what the DM target did. For BIOs that are issued by the
user/FS to the DM device, then that is when bio_needs_zone_write_plugging() is
used by DM core to determine if a BIO must be split.

Anyway, I do not think you need to reference this commit. So I would drop this
entire paragraph from the commit message, and start the explanation with the
below paragraph. That is very clear I think.

> Commit 2df7168717b7 ("dm: Always split write BIOs to zoned device
> limits") updates the device-mapper driver to perform splits for the
> write BIOs. However, it did not address the cases where DM targets do
> not emulate zone append, such as in the cases of dm-linear or dm-flakey.
> For these targets, when the write BIOs span across zone boundaries, they
> trigger WARN_ON_ONCE(bio_straddles_zones(bio)) in
> blk_zone_wplug_handle_write(). This results in I/O errors. The errors
> are reproduced by running blktests test case zbd/004 using zoned
> dm-linear or dm-flakey devices.
> 
> To avoid the I/O errors, handle the write BIOs regardless whether DM
> targets emulate zone append or not, so that all write BIOs are split at
> zone boundaries. For that purpose, drop the check for zone append
> emulation in dm_zone_bio_needs_split(). Its argument 'md' is no longer
> used then drop it also.
> 
> Fixes: 2df7168717b7 ("dm: Always split write BIOs to zoned device limits")
> Cc: stable@vger.kernel.org

The above patch is queued in block-next, so there is no need for this CC-stable.
Remove it please.

> Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>

One more nit below.

With these fixed, feel free to add:

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

> ---
>  drivers/md/dm.c | 17 ++++++-----------
>  1 file changed, 6 insertions(+), 11 deletions(-)
> 
> diff --git a/drivers/md/dm.c b/drivers/md/dm.c
> index ca889328fdfe..a64809f04241 100644
> --- a/drivers/md/dm.c
> +++ b/drivers/md/dm.c
> @@ -1785,8 +1785,7 @@ static void init_clone_info(struct clone_info *ci, struct dm_io *io,
>  }
>  
>  #ifdef CONFIG_BLK_DEV_ZONED
> -static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
> -					   struct bio *bio)
> +static inline bool dm_zone_bio_needs_split(struct bio *bio)
>  {
>  	/*
>  	 * Special case the zone operations that cannot or should not be split.
> @@ -1802,13 +1801,10 @@ static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
>  	}
>  
>  	/*
> -	 * Mapped devices that require zone append emulation will use the block
> -	 * layer zone write plugging. In such case, we must split any large BIO
> -	 * to the mapped device limits to avoid potential deadlocks with queue
> -	 * freeze operations.
> +	 * When mapped devices use the block layer zone write plugging, we must
> +	 * split any large BIO to the mapped device limits to avoid potential
> +	 * deadlocks with queue freeze operations.

Nit:
	 * split any large BIO to the mapped device limits to not submit BIOs
	 * that span zone boundaries and to avoid potential deadlocks with
	 * queue freeze operations.


>  	 */
> -	if (!dm_emulate_zone_append(md))
> -		return false;
>  	return bio_needs_zone_write_plugging(bio) || bio_straddles_zones(bio);
>  }
>  
> @@ -1932,8 +1928,7 @@ static blk_status_t __send_zone_reset_all(struct clone_info *ci)
>  }
>  
>  #else
> -static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
> -					   struct bio *bio)
> +static inline bool dm_zone_bio_needs_split(struct bio *bio)
>  {
>  	return false;
>  }
> @@ -1960,7 +1955,7 @@ static void dm_split_and_process_bio(struct mapped_device *md,
>  
>  	is_abnormal = is_abnormal_io(bio);
>  	if (static_branch_unlikely(&zoned_enabled)) {
> -		need_split = is_abnormal || dm_zone_bio_needs_split(md, bio);
> +		need_split = is_abnormal || dm_zone_bio_needs_split(bio);
>  	} else {
>  		need_split = is_abnormal;
>  	}


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH for-next] dm: split write BIOs on zone boundaries when zone append is not emulated

[Shinichiro Kawasaki]

On Jul 17, 2025 / 17:18, Damien Le Moal wrote:
> On 2025/07/17 11:32, Shin'ichiro Kawasaki wrote:
> > Since commit f70291411ba2 ("block: Introduce
> > bio_needs_zone_write_plugging()"), blk_mq_submit_bio() no longer splits
> > write BIOs to comply with the requirements of zoned block devices. This
> > change applies to write BIOs issued by zoned DM targets. Following this
> > modification, it is expected that DM targets themselves split write BIOs
> > for zoned block devices.
> 
> This description is confusing: if the underlying device of a DM target is a
> blk-mq block device, then blk_mq_submit_bio() will be called and will split BIOs
> as needed, regardless of what the DM target did. For BIOs that are issued by the
> user/FS to the DM device, then that is when bio_needs_zone_write_plugging() is
> used by DM core to determine if a BIO must be split.
> 
> Anyway, I do not think you need to reference this commit. So I would drop this
> entire paragraph from the commit message, and start the explanation with the
> below paragraph. That is very clear I think.
> 
> > Commit 2df7168717b7 ("dm: Always split write BIOs to zoned device
> > limits") updates the device-mapper driver to perform splits for the
> > write BIOs. However, it did not address the cases where DM targets do
> > not emulate zone append, such as in the cases of dm-linear or dm-flakey.
> > For these targets, when the write BIOs span across zone boundaries, they
> > trigger WARN_ON_ONCE(bio_straddles_zones(bio)) in
> > blk_zone_wplug_handle_write(). This results in I/O errors. The errors
> > are reproduced by running blktests test case zbd/004 using zoned
> > dm-linear or dm-flakey devices.
> > 
> > To avoid the I/O errors, handle the write BIOs regardless whether DM
> > targets emulate zone append or not, so that all write BIOs are split at
> > zone boundaries. For that purpose, drop the check for zone append
> > emulation in dm_zone_bio_needs_split(). Its argument 'md' is no longer
> > used then drop it also.
> > 
> > Fixes: 2df7168717b7 ("dm: Always split write BIOs to zoned device limits")
> > Cc: stable@vger.kernel.org
> 
> The above patch is queued in block-next, so there is no need for this CC-stable.
> Remove it please.
> 
> > Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
> 
> One more nit below.
> 
> With these fixed, feel free to add:
> 
> Reviewed-by: Damien Le Moal <dlemoal@kernel.org>

Thanks. I will reflect your comments and send out v2 soon.