diff options
| author | Jaegeuk Kim <jaegeuk@kernel.org> | 2014-09-10 15:04:03 -0700 |
|---|---|---|
| committer | Jaegeuk Kim <jaegeuk@kernel.org> | 2014-09-23 11:10:16 -0700 |
| commit | 19c9c466e5c79402cbf8929f922011b02fbb59bd (patch) | |
| tree | 632f1646864c395780dc360add826f1f24e33529 /fs | |
| parent | 441ac5cb323a47b0a665f77f7cd6f76aacbdf21c (diff) | |
| download | op-kernel-dev-19c9c466e5c79402cbf8929f922011b02fbb59bd.zip op-kernel-dev-19c9c466e5c79402cbf8929f922011b02fbb59bd.tar.gz | |
f2fs: do not skip latest inode information
In f2fs_sync_file, if there is no written appended writes, it skips
to write its node blocks.
But, if there is up-to-date inode page, we should write it to update
its metadata during the roll-forward recovery.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Diffstat (limited to 'fs')
| -rw-r--r-- | fs/f2fs/file.c | 11 |
1 files changed, 10 insertions, 1 deletions
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 3035c79..a041c66 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -169,12 +169,21 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) */ if (!is_inode_flag_set(fi, FI_APPEND_WRITE) && !exist_written_data(sbi, ino, APPEND_INO)) { + struct page *i = find_get_page(NODE_MAPPING(sbi), ino); + + /* But we need to avoid that there are some inode updates */ + if ((i && PageDirty(i)) || need_inode_block_update(sbi, ino)) { + f2fs_put_page(i, 0); + goto go_write; + } + f2fs_put_page(i, 0); + if (is_inode_flag_set(fi, FI_UPDATE_WRITE) || exist_written_data(sbi, ino, UPDATE_INO)) goto flush_out; goto out; } - +go_write: /* guarantee free sections for fsync */ f2fs_balance_fs(sbi); |
