When fsync() system call is executed, all data of the file should be written on the storage.
But, in a case of that the file is operated in data journal mode, the data doesn't need to be
flushed out and waited for its I/O completion. Moreover, by invoking filemap_fdatawrite()
function on the data journaled file, ext4_sync_file() suffers from an unnecessary delay
which is caused by flushing out and waiting for the I/O completion of a ton of newly-dirtied
pages which were dirtied by the previous commit.
Signed-off-by: Daeho Jeong <daeho.jeong@xxxxxxxxxxx>
---
fs/ext4/fsync.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index 8850254..e0fabf7 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -112,9 +112,6 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
goto out;
}
- ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
- if (ret)
- return ret;
/*
* data=writeback,ordered:
* The caller's filemap_fdatawrite()/wait will sync the data.
@@ -125,13 +122,15 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
* filemap_fdatawrite won't do anything (the buffers are clean).
* ext4_force_commit will write the file data into the journal and
* will wait on that.
- * filemap_fdatawait() will encounter a ton of newly-dirtied pages
- * (they were dirtied by commit). But that's OK - the blocks are
- * safe in-journal, which is all fsync() needs to ensure.
*/
if (ext4_should_journal_data(inode)) {
ret = ext4_force_commit(inode->i_sb);
goto out;
+ } else {
+ ret = filemap_write_and_wait_range(inode->i_mapping,
+ start, end);
+ if (ret)
+ return ret;
}
commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid;
--
1.7.9.5��.n��������+%������w��{.n�����{�����ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
