Re: [PATCH V2] f2fs: fix to adjust appropriate length for fiemap

[Chao Yu <chao@kernel.org>]

On 2024/11/5 12:02, Zhiguo Niu wrote:
> Chao Yu <chao@kernel.org> 于2024年11月5日周二 11:15写道：
>>
>> On 2024/11/4 9:56, Zhiguo Niu wrote:
>>> If user give a file size as "length" parameter for fiemap
>>> operations, but if this size is non-block size aligned,
>>> it will show 2 segments fiemap results even this whole file
>>> is contiguous on disk, such as the following results:
>>>
>>>    ./f2fs_io fiemap 0 19034 ylog/analyzer.py
>>> Fiemap: offset = 0 len = 19034
>>>           logical addr.    physical addr.   length           flags
>>> 0       0000000000000000 0000000020baa000 0000000000004000 00001000
>>> 1       0000000000004000 0000000020bae000 0000000000001000 00001001
>>>
>>> after this patch:
>>> ./f2fs_io fiemap 0 19034 ylog/analyzer.py
>>> Fiemap: offset = 0 len = 19034
>>>       logical addr.    physical addr.   length           flags
>>> 0    0000000000000000 00000000315f3000 0000000000005000 00001001
>>
>> Hi Zhiguo,
>>
>> Any testcase to reproduce this bug? w/o this patch, it looks output
>> from fiemap looks fine?
>>
>> f2fs_io fiemap 0 19034 file
>> Fiemap: offset = 0 len = 19034
>>          logical addr.    physical addr.   length           flags
>> 0       0000000000000000 0000000004401000 0000000000005000 00001001
>>
> Hi Chao,
> Sorry I didn't write clearly enough about the test case, and i put the
> note below the ""Singed-off" tag.
> let me describe it again， f2fs_io fiemap has been modified by me for
> testing in my local, and  the length parameter
> is the real file size of the file, not the block numer. because user
> also pass the real file size to fiemap ioctl.
> so with the new f2fs_io fiemap, a contiguous file on disk may be shown
> 2 segments if the length is not block size alinged.
> such as:
> 
> ums9632_1h10:/data # ls -l ylog/ap/analyzer.py
> -rw-rw-rw- 1 root system 19006 2008-01-01 00:00 ylog/ap/analyzer.py
> ums9632_1h10:/data # ./f2fs_io fiemap 0 19006 ylog/ap/analyzer.py
> Fiemap: offset = 0 len = 19006
>          logical addr.    physical addr.   length           flags
> 0       0000000000000000 0000000020baa000 0000000000004000 00001000
> 1       0000000000004000 0000000020bae000 0000000000001000 00001001
> 
> but if we pass a length that is block size alinged, it will show one
> whole segment in fiemap log.
> ums9632_1h10:/data # ./f2fs_io fiemap 0 20480 ylog/ap/analyzer.py
> Fiemap: offset = 0 len = 20480
>          logical addr.    physical addr.   length           flags
> 0       0000000000000000 0000000020baa000 0000000000005000 00001001
> Thanks.

Oh, I can reproduce it w/ xfs_io, thanks for your explanation.

xfs_io file -c "fiemap -v 0 19006"
file:
  EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
    0: [0..31]:         139272..139303      32 0x1000
    1: [32..39]:        139304..139311       8 0x1001
xfs_io file -c "fiemap -v 0 20480"
file:
  EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
    0: [0..39]:         139272..139311      40 0x1001

Thanks,

>> Thanks,
>>
>>>
>>> Signed-off-by: Zhiguo Niu <zhiguo.niu@unisoc.com>
>>> ---
>>> V2: correct commit msg according to Chao's questions
>>> f2fs_io has been modified for testing, the length for fiemap is
>>> real file size, not block number
>>> ---
>>>    fs/f2fs/data.c | 4 ++--
>>>    1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
>>> index 306b86b0..9fc229d 100644
>>> --- a/fs/f2fs/data.c
>>> +++ b/fs/f2fs/data.c
>>> @@ -1966,8 +1966,8 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
>>>                        goto out;
>>>        }
>>>
>>> -     if (bytes_to_blks(inode, len) == 0)
>>> -             len = blks_to_bytes(inode, 1);
>>> +     if (len & (blks_to_bytes(inode, 1) - 1))
>>> +             len = round_up(len, blks_to_bytes(inode, 1));
>>>
>>>        start_blk = bytes_to_blks(inode, start);
>>>        last_blk = bytes_to_blks(inode, start + len - 1);
>>