Re: [PATCH v7 01/16] fs: Add case sensitivity flags to file_kattr

["Chuck Lever" <cel@kernel.org>]

On Thu, Jan 22, 2026, at 7:26 PM, Darrick J. Wong wrote:
> On Thu, Jan 22, 2026 at 11:02:56AM -0500, Chuck Lever wrote:
>> From: Chuck Lever <chuck.lever@oracle.com>
>> 
>> Enable upper layers such as NFSD to retrieve case sensitivity
>> information from file systems by adding FS_XFLAG_CASEFOLD and
>> FS_XFLAG_CASENONPRESERVING flags.
>> 
>> Filesystems report case-insensitive or case-nonpreserving behavior
>> by setting these flags directly in fa->fsx_xflags. The default
>> (flags unset) indicates POSIX semantics: case-sensitive and
>> case-preserving. These flags are read-only; userspace cannot set
>> them via ioctl.
>> 
>> Remove struct file_kattr initialization from fileattr_fill_xflags()
>> and fileattr_fill_flags(). Callers at ioctl/syscall entry points
>> zero-initialize the struct themselves, which allows them to pass
>> hints (flags_valid, fsx_valid) to the filesystem's ->fileattr_get()
>> callback via the fa argument. Filesystem handlers that invoke these
>> fill functions can now set flags directly in fa->fsx_xflags before
>> calling them, without the fill functions zeroing those values.
>
> In hindsight I regret not asking for the file_kattr initialization
> change to be in a separate patch.

If I’m asked to post another revision of this series, I’ll look at splitting
1/16 into two patches.


>> Case sensitivity information is exported to userspace via the
>> fa_xflags field in the FS_IOC_FSGETXATTR ioctl and file_getattr()
>> system call.
>> 
>> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
>
> The UAPI changes still look ok to me.  AFAICT the file_kattr
> initialization now seem like they don't zap fields to confuse
> vfs_fileattr_get.
>
> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>


-- 
Chuck Lever