2022-11-17 01:27:50

by Ye Bin

Subject: [PATCH 2/3] ext4: WANR_ON when detect abnormal 'i_reserved_data_blocks'

From: Ye Bin <[email protected]>

If 'i_reserved_data_blocks' is not cleared which mean something wrong
with code, so emit WARN_ON to capture this abnormal closer to the first
scene.

Signed-off-by: Ye Bin <[email protected]>
---
fs/ext4/super.c | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 63ef74eb8091..30885a6fe18b 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1385,11 +1385,14 @@ static void ext4_destroy_inode(struct inode *inode)
dump_stack();
}

- if (EXT4_I(inode)->i_reserved_data_blocks)
- ext4_msg(inode->i_sb, KERN_ERR,
- "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",
- inode->i_ino, EXT4_I(inode),
- EXT4_I(inode)->i_reserved_data_blocks);
+ if (EXT4_I(inode)->i_reserved_data_blocks) {
+ ext4_warning(inode->i_sb, "Inode %lu (%p): "
+ "i_reserved_data_blocks (%u) not cleared!",
+ inode->i_ino, EXT4_I(inode),
+ EXT4_I(inode)->i_reserved_data_blocks);
+
+ WARN_ON(1);
+ }
}

static void init_once(void *foo)
--
2.31.1

2022-11-21 09:50:29

by Jan Kara

Subject: Re: [PATCH 2/3] ext4: WANR_ON when detect abnormal 'i_reserved_data_blocks'

On Thu 17-11-22 09:42:45, Ye Bin wrote:
> From: Ye Bin <[email protected]>
>
> If 'i_reserved_data_blocks' is not cleared which mean something wrong
> with code, so emit WARN_ON to capture this abnormal closer to the first
> scene.
>
> Signed-off-by: Ye Bin <[email protected]>
> ---
> fs/ext4/super.c | 13 ++++++++-----
> 1 file changed, 8 insertions(+), 5 deletions(-)
>
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index 63ef74eb8091..30885a6fe18b 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -1385,11 +1385,14 @@ static void ext4_destroy_inode(struct inode *inode)
> dump_stack();
> }
>
> - if (EXT4_I(inode)->i_reserved_data_blocks)
> - ext4_msg(inode->i_sb, KERN_ERR,
> - "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",
> - inode->i_ino, EXT4_I(inode),
> - EXT4_I(inode)->i_reserved_data_blocks);
> + if (EXT4_I(inode)->i_reserved_data_blocks) {
> + ext4_warning(inode->i_sb, "Inode %lu (%p): "
> + "i_reserved_data_blocks (%u) not cleared!",
> + inode->i_ino, EXT4_I(inode),
> + EXT4_I(inode)->i_reserved_data_blocks);
> +
> + WARN_ON(1);
> + }

Hum, so I'd think that if this happens, the free space accounting is likely
wrong so we might as well just force the filesystem to error mode with
ext4_error() to force fsck? I also gives a good chance to various test
systems to detect there is some problem so we don't need the WARN_ON then?
What do others think?

Honza
--
Jan Kara <[email protected]>
SUSE Labs, CR

2022-11-21 11:25:43

by Ye Bin

Subject: Re: [PATCH 2/3] ext4: WANR_ON when detect abnormal 'i_reserved_data_blocks'

On 2022/11/21 17:47, Jan Kara wrote:
> On Thu 17-11-22 09:42:45, Ye Bin wrote:
>> From: Ye Bin <[email protected]>
>>
>> If 'i_reserved_data_blocks' is not cleared which mean something wrong
>> with code, so emit WARN_ON to capture this abnormal closer to the first
>> scene.
>>
>> Signed-off-by: Ye Bin <[email protected]>
>> ---
>> fs/ext4/super.c | 13 ++++++++-----
>> 1 file changed, 8 insertions(+), 5 deletions(-)
>>
>> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
>> index 63ef74eb8091..30885a6fe18b 100644
>> --- a/fs/ext4/super.c
>> +++ b/fs/ext4/super.c
>> @@ -1385,11 +1385,14 @@ static void ext4_destroy_inode(struct inode *inode)
>> dump_stack();
>> }
>>
>> - if (EXT4_I(inode)->i_reserved_data_blocks)
>> - ext4_msg(inode->i_sb, KERN_ERR,
>> - "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",
>> - inode->i_ino, EXT4_I(inode),
>> - EXT4_I(inode)->i_reserved_data_blocks);
>> + if (EXT4_I(inode)->i_reserved_data_blocks) {
>> + ext4_warning(inode->i_sb, "Inode %lu (%p): "
>> + "i_reserved_data_blocks (%u) not cleared!",
>> + inode->i_ino, EXT4_I(inode),
>> + EXT4_I(inode)->i_reserved_data_blocks);
>> +
>> + WARN_ON(1);
>> + }
> Hum, so I'd think that if this happens, the free space accounting is likely
> wrong so we might as well just force the filesystem to error mode with
> ext4_error() to force fsck? I also gives a good chance to various test
> systems to detect there is some problem so we don't need the WARN_ON then?
> What do others think?
>
> Honza
Thanks for your advice, use ext4_error() maybe is suitable and also
testable.