linux.git/fs/ext4/inode.c, branch v6.10

ext4: remove the redundant folio_wait_stable()

2024-05-07T19:48:04+00:00

__filemap_get_folio() with FGP_WRITEBEGIN parameter has already wait
for stable folio, so remove the redundant folio_wait_stable() in
ext4_da_write_begin(), it was left over from the commit cc883236b792
("ext4: drop unnecessary journal handle in delalloc write") that
removed the retry getting page logic.

Fixes: cc883236b792 ("ext4: drop unnecessary journal handle in delalloc write")
Signed-off-by: Zhang Yi 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/20240419023005.2719050-1-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o

ext4: set FMODE_CAN_ODIRECT instead of a dummy direct_IO method

2024-05-02T14:53:32+00:00

Since commit a2ad63daa88b ("VFS: add FMODE_CAN_ODIRECT file flag") file
systems can just set the FMODE_CAN_ODIRECT flag at open time instead of
wiring up a dummy direct_IO method to indicate support for direct I/O.

Signed-off-by: Christoph Hellwig 
[RH: Rebased to upstream]
Signed-off-by: Ritesh Harjani (IBM) 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/e5797bb597219a49043e53e4e90aa494b97dc328.1709215665.git.ritesh.list@gmail.com
Signed-off-by: Theodore Ts'o

ext4: Remove PAGE_MASK dependency on mpage_submit_folio

2024-05-02T14:50:44+00:00

This patch simply removes the PAGE_MASK dependency since
mpage_submit_folio() is already converted to work with folio.

Signed-off-by: Ritesh Harjani (IBM) 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/d6eadb090334ea49ceef4e643b371fabfcea328f.1709182251.git.ritesh.list@gmail.com
Signed-off-by: Theodore Ts'o

ext4: Fixes len calculation in mpage_journal_page_buffers

2024-05-02T14:50:44+00:00

Truncate operation can race with writeback, in which inode->i_size can get
truncated and therefore size - folio_pos() can be negative. This fixes the
len calculation. However this path doesn't get easily triggered even
with data journaling.

Cc: stable@kernel.org # v6.5
Fixes: 80be8c5cc925 ("Fixes: ext4: Make mpage_journal_page_buffers use folio")
Signed-off-by: Ritesh Harjani (IBM) 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/cff4953b5c9306aba71e944ab176a5d396b9a1b7.1709182250.git.ritesh.list@gmail.com
Signed-off-by: Theodore Ts'o

ext4: fix the comment of ext4_map_blocks()/ext4_ext_map_blocks()

2024-02-22T03:33:21+00:00

this comment of ext4_map_blocks()/ext4_ext_map_blocks() need
update after commit c21770573319("ext4: Define a new set of
flags for ext4_get_blocks()").

Signed-off-by: Cheng Nie 
Link: https://lore.kernel.org/r/20240118062511.28276-1-niecheng1@uniontech.com
Signed-off-by: Theodore Ts'o

ext4: make ext4_set_iomap() recognize IOMAP_DELALLOC map type

2024-02-02T04:59:21+00:00

Since ext4_map_blocks() can recognize a delayed allocated only extent,
make ext4_set_iomap() can also recognize it, and remove the useless
separate check in ext4_iomap_begin_report().

Signed-off-by: Zhang Yi 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/20240127015825.1608160-7-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o

ext4: make ext4_map_blocks() distinguish delalloc only extent

2024-02-02T04:59:21+00:00

Add a new map flag EXT4_MAP_DELAYED to indicate the mapping range is a
delayed allocated only (not unwritten) one, and making
ext4_map_blocks() can distinguish it, no longer mixing it with holes.

Signed-off-by: Zhang Yi 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/20240127015825.1608160-6-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o

ext4: add a hole extent entry in cache after punch

2024-02-02T04:59:21+00:00

In order to cache hole extents in the extent status tree and keep the
hole length as long as possible, re-add a hole entry to the cache just
after punching a hole.

Signed-off-by: Zhang Yi 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/20240127015825.1608160-5-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o

ext4: convert to exclusive lock while inserting delalloc extents

2024-02-02T04:47:02+00:00

ext4_da_map_blocks() only hold i_data_sem in shared mode and i_rwsem
when inserting delalloc extents, it could be raced by another querying
path of ext4_map_blocks() without i_rwsem, .e.g buffered read path.
Suppose we buffered read a file containing just a hole, and without any
cached extents tree, then it is raced by another delayed buffered write
to the same area or the near area belongs to the same hole, and the new
delalloc extent could be overwritten to a hole extent.

 pread()                           pwrite()
  filemap_read_folio()
   ext4_mpage_readpages()
    ext4_map_blocks()
     down_read(i_data_sem)
     ext4_ext_determine_hole()
     //find hole
     ext4_ext_put_gap_in_cache()
      ext4_es_find_extent_range()
      //no delalloc extent
                                    ext4_da_map_blocks()
                                     down_read(i_data_sem)
                                     ext4_insert_delayed_block()
                                     //insert delalloc extent
      ext4_es_insert_extent()
      //overwrite delalloc extent to hole

This race could lead to inconsistent delalloc extents tree and
incorrect reserved space counter. Fix this by converting to hold
i_data_sem in exclusive mode when adding a new delalloc extent in
ext4_da_map_blocks().

Cc: stable@vger.kernel.org
Signed-off-by: Zhang Yi 
Suggested-by: Jan Kara 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/20240127015825.1608160-3-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o

ext4: refactor ext4_da_map_blocks()

2024-02-02T04:47:02+00:00

Refactor and cleanup ext4_da_map_blocks(), reduce some unnecessary
parameters and branches, no logic changes.

Signed-off-by: Zhang Yi 
Reviewed-by: Jan Kara 
Link: https://lore.kernel.org/r/20240127015825.1608160-2-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o