Section: Linux Programmer's Manual (2)
Return to Main Contents
fsync, fdatasync - synchronize a file's in-core state with storage device
int fsync(int fd);
int fdatasync(int fd);
Feature Test Macro Requirements for glibc (see
_BSD_SOURCE || _XOPEN_SOURCE
|| /* since glibc 2.8: */ _POSIX_C_SOURCE >= 200112L
_POSIX_C_SOURCE >= 199309L || _XOPEN_SOURCE >= 500
transfers ("flushes") all modified in-core data of
(i.e., modified buffer cache pages for) the
file referred to by the file descriptor
to the disk device (or other permanent storage device)
where that file resides.
The call blocks until the device reports that the transfer has completed.
It also flushes metadata information associated with the file (see
does not necessarily ensure
that the entry in the directory containing the file has also reached disk.
For that an explicit
on a file descriptor for the directory is also needed.
is similar to
but does not flush modified metadata unless that metadata
is needed in order to allow a subsequent data retrieval to be
For example, changes to
(respectively, time of last access and
time of last modification; see
do not require flushing because they are not necessary for
a subsequent data read to be handled correctly.
On the other hand, a change to the file size
as made by say
would require a metadata flush.
The aim of
is to reduce disk activity for applications that do not
require all metadata to be synchronized with the disk.
On success, these system calls return zero.
On error, -1 is returned, and
is set appropriately.
is not a valid file descriptor open for writing.
An error occurred during synchronization.
- EROFS, EINVAL
is bound to a special file which does not support synchronization.
On POSIX systems on which
is defined in
to a value greater than 0.
Applications that access databases or log files often write a tiny
data fragment (e.g., one line in a log file) and then call
immediately in order to ensure that the written data is physically
stored on the harddisk.
will always initiate two write operations: one for the newly written
data and another one in order to update the modification time stored
in the inode.
If the modification time is not a part of the transaction
can be used to avoid unnecessary inode disk write operations.
If the underlying hard disk has write caching enabled, then
the data may not really be on permanent storage when
When an ext2 file system is mounted with the
option, directory entries are also implicitly synced by
On kernels before 2.4,
on big files can be inefficient.
An alternative might be to use the
In Linux 2.2 and earlier,
is equivalent to
and so has no performance advantage.
This page is part of release 3.27 of the Linux
A description of the project,
and information about reporting bugs,
can be found at
- RETURN VALUE
- CONFORMING TO
- SEE ALSO
This document was created by
using the manual pages.
Time: 19:49:18 GMT, April 27, 2011