[PATCH] btrfs: zoned: propagate do_zone_finish() error in btrfs_zone_finish_endio()

[PATCH] btrfs: zoned: propagate do_zone_finish() error in btrfs_zone_finish_endio()

[Johannes Thumshirn]

btrfs_zone_finish_endio() ignored the return value of do_zone_finish()
and always returned 0, silently dropping a failed zone finish.

Instead propagate any error from do_zone_finish() as the caller
btrfs_finish_ordered_io() already handles it.

Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
---
 fs/btrfs/zoned.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index a016cb471beb..82c2394dd43c 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -2713,6 +2713,7 @@ int btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical, u64 leng
 {
 	struct btrfs_block_group *block_group;
 	u64 min_alloc_bytes;
+	int ret = 0;
 
 	if (!btrfs_is_zoned(fs_info))
 		return 0;
@@ -2732,11 +2733,11 @@ int btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical, u64 leng
 	    block_group->start + block_group->zone_capacity)
 		goto out;
 
-	do_zone_finish(block_group, true);
+	ret = do_zone_finish(block_group, true);
 
 out:
 	btrfs_put_block_group(block_group);
-	return 0;
+	return ret;
 }
 
 static void btrfs_zone_finish_endio_workfn(struct work_struct *work)
-- 
2.55.0

Re: [PATCH] btrfs: zoned: propagate do_zone_finish() error in btrfs_zone_finish_endio()

[Qu Wenruo]

在 2026/8/18 19:30, Johannes Thumshirn 写道:
> btrfs_zone_finish_endio() ignored the return value of do_zone_finish()
> and always returned 0, silently dropping a failed zone finish.
> 
> Instead propagate any error from do_zone_finish() as the caller
> btrfs_finish_ordered_io() already handles it.
> 
> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Reviewed-by: Qu Wenruo <wqu@suse.com>

Thanks,
Qu

> ---
>   fs/btrfs/zoned.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
> index a016cb471beb..82c2394dd43c 100644
> --- a/fs/btrfs/zoned.c
> +++ b/fs/btrfs/zoned.c
> @@ -2713,6 +2713,7 @@ int btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical, u64 leng
>   {
>   	struct btrfs_block_group *block_group;
>   	u64 min_alloc_bytes;
> +	int ret = 0;
>   
>   	if (!btrfs_is_zoned(fs_info))
>   		return 0;
> @@ -2732,11 +2733,11 @@ int btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical, u64 leng
>   	    block_group->start + block_group->zone_capacity)
>   		goto out;
>   
> -	do_zone_finish(block_group, true);
> +	ret = do_zone_finish(block_group, true);
>   
>   out:
>   	btrfs_put_block_group(block_group);
> -	return 0;
> +	return ret;
>   }
>   
>   static void btrfs_zone_finish_endio_workfn(struct work_struct *work)

Re: [PATCH] btrfs: zoned: propagate do_zone_finish() error in btrfs_zone_finish_endio()

[Qu Wenruo]

在 2026/8/19 08:11, Qu Wenruo 写道:
> 
> 
> 在 2026/8/18 19:30, Johannes Thumshirn 写道:
>> btrfs_zone_finish_endio() ignored the return value of do_zone_finish()
>> and always returned 0, silently dropping a failed zone finish.
>>
>> Instead propagate any error from do_zone_finish() as the caller
>> btrfs_finish_ordered_io() already handles it.
>>
>> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
> 
> Reviewed-by: Qu Wenruo <wqu@suse.com>

BTW, you may still want to check the sashiko review, which exposed an 
pre-existing bug in the error handling path:

https://sashiko.dev/#/patchset/20260818100037.1366563-1-johannes.thumshirn%40wdc.com

> 
> Thanks,
> Qu
> 
>> ---
>>   fs/btrfs/zoned.c | 5 +++--
>>   1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
>> index a016cb471beb..82c2394dd43c 100644
>> --- a/fs/btrfs/zoned.c
>> +++ b/fs/btrfs/zoned.c
>> @@ -2713,6 +2713,7 @@ int btrfs_zone_finish_endio(struct btrfs_fs_info 
>> *fs_info, u64 logical, u64 leng
>>   {
>>       struct btrfs_block_group *block_group;
>>       u64 min_alloc_bytes;
>> +    int ret = 0;
>>       if (!btrfs_is_zoned(fs_info))
>>           return 0;
>> @@ -2732,11 +2733,11 @@ int btrfs_zone_finish_endio(struct 
>> btrfs_fs_info *fs_info, u64 logical, u64 leng
>>           block_group->start + block_group->zone_capacity)
>>           goto out;
>> -    do_zone_finish(block_group, true);
>> +    ret = do_zone_finish(block_group, true);
>>   out:
>>       btrfs_put_block_group(block_group);
>> -    return 0;
>> +    return ret;
>>   }
>>   static void btrfs_zone_finish_endio_workfn(struct work_struct *work)
> 
> 

Re: [PATCH] btrfs: zoned: propagate do_zone_finish() error in btrfs_zone_finish_endio()

[Johannes Thumshirn]

On 8/19/26 12:44 AM, Qu Wenruo wrote:
>
>
> 在 2026/8/19 08:11, Qu Wenruo 写道:
>>
>>
>> 在 2026/8/18 19:30, Johannes Thumshirn 写道:
>>> btrfs_zone_finish_endio() ignored the return value of do_zone_finish()
>>> and always returned 0, silently dropping a failed zone finish.
>>>
>>> Instead propagate any error from do_zone_finish() as the caller
>>> btrfs_finish_ordered_io() already handles it.
>>>
>>> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
>>
>> Reviewed-by: Qu Wenruo <wqu@suse.com>
>
> BTW, you may still want to check the sashiko review, which exposed an 
> pre-existing bug in the error handling path:
>
> https://sashiko.dev/#/patchset/20260818100037.1366563-1-johannes.thumshirn%40wdc.com


Oh thanks, I would've missed this. From a quick glance it looks valid. 
I'll take care of it.