Re: [PATCH] ext4: prevent inline data creation on extent-based inodes

["Zhou, Yun" <yun.zhou@windriver.com>]

On 7/2/26 09:38, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
> 
> New issues:
> - [High] The patch completely breaks the inline data feature for all new files and directories on extents-enabled filesystems.
> 
> Pre-existing issues:
> - [High] The patch fails to resolve the data corruption race condition on filesystems where the `extents` feature is disabled, leaving indirect-mapped files vulnerable.
> --
> 
>> diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
>> index 0eff4a0c6a6c3..2fcafbdb7b15c 100644
>> --- a/fs/ext4/inline.c
>> +++ b/fs/ext4/inline.c
>> @@ -418,6 +418,13 @@ static int ext4_prepare_inline_data(handle_t *handle, struct inode *inode,
>>                return -ENOSPC;
>>
>>        ext4_write_lock_xattr(inode, &no_expand);
>> +
>> +     /* Once converted to extents, never go back to inline data. */
>> +     if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
> 
> [Severity: High]
> Does this completely disable inline data for all newly created files and
> directories on extents-enabled filesystems?
> 
> Looking at __ext4_new_inode(), it appears newly created inodes inherently
> have the EXT4_INODE_EXTENTS flag set:
> 
> fs/ext4/ialloc.c:__ext4_new_inode() {
>      ...
>      ext4_set_inode_flag(inode, EXT4_INODE_EXTENTS);
>      ...
> }
> 
> If ext4_prepare_inline_data() is called during a write or directory creation
> for a new inode, wouldn't this new check evaluate to true and unconditionally
> abort the inline data creation by returning -ENOSPC?
> 
My apologies, my analysis was incorrect, and I am currently working on a 
new fix.

Please ignore this version.