]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
ext4: Do not fallback to buffered-io for DIO atomic write
authorRitesh Harjani (IBM) <ritesh.list@gmail.com>
Tue, 5 Nov 2024 00:22:59 +0000 (16:22 -0800)
committerDarrick J. Wong <djwong@djwong.org>
Wed, 6 Nov 2024 00:20:40 +0000 (16:20 -0800)
commit299537e9dfac2ecd08e7dae87a6437b92612568a
tree9a561a97b42f4d6145037af0add3994fc45dae18
parentb7987a7d69a4a17b7f334f5c100d6729ebcc2ccb
ext4: Do not fallback to buffered-io for DIO atomic write

atomic writes is currently only supported for single fsblock and only
for direct-io. We should not return -ENOTBLK for atomic writes since we
want the atomic write request to either complete fully or fail
otherwise. Hence, we should never fallback to buffered-io in case of
DIO atomic write requests.
Let's also catch if this ever happens by adding some WARN_ON_ONCE before
buffered-io handling for direct-io atomic writes. More details of the
discussion [1].

While at it let's add an inline helper ext4_want_directio_fallback() which
simplifies the logic checks and inherently fixes condition on when to return
-ENOTBLK which otherwise was always returning true for any write or directio in
ext4_iomap_end(). It was ok since ext4 only supports direct-io via iomap.

[1]: https://lore.kernel.org/linux-xfs/cover.1729825985.git.ritesh.list@gmail.com/T/#m9dbecc11bed713ed0d7a486432c56b105b555f04
Suggested-by: Darrick J. Wong <djwong@kernel.org> # inline helper
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
fs/ext4/file.c
fs/ext4/inode.c