stat() is a Unix system call that returns useful data about a file inode. The semantics of
stat() vary between operating systems. As an example, the Unix command ls uses it to retrieve information on (among many others):
stat() functions and stat structureEdit
int stat(const char *filename, struct stat *buf); int lstat(const char *filename, struct stat *buf); int fstat(int filedesc, struct stat *buf);
and defines the
struct stat structure as including at least the following members:
dev_t st_dev; /* ID of device containing file */ ino_t st_ino; /* inode number */ mode_t st_mode; /* protection */ nlink_t st_nlink; /* number of hard links */ uid_t st_uid; /* user ID of owner */ gid_t st_gid; /* group ID of owner */ dev_t st_rdev; /* device ID (if special file) */ off_t st_size; /* total size, in bytes */ time_t st_atime; /* time of last access */ time_t st_mtime; /* time of last modification */ time_t st_ctime; /* time of last status change */ blksize_t st_blksize; /* blocksize for filesystem I/O */ blkcnt_t st_blocks; /* number of blocks allocated */
lstat() is a library function that retrieves the status of a file. It is identical to
stat(), except when the file is a symbolic link, in which case information about the link itself is returned instead of the linked-to file.
Criticism of atimeEdit
Writing to a file changes its
ctime, while reading a file changes its
atime. As a result, on a POSIX-compliant system, reading a file causes a write, which has been criticized. This behaviour can usually be disabled by adding a mount option in /etc/fstab.
However, turning off atime updating breaks POSIX compliance, and some applications, notably the mutt mail reader (in some configurations), and some file usage watching utilities, notably tmpwatch. In the worst case, not updating atime can cause some backup programs to fail to backup a file.
Linux kernel developer Ingo Molnár called atime "perhaps the most stupid Unix design idea of all times," adding: "[T]hink about this a bit: 'For every file that is read from the disk, lets do a ... write to the disk! And, for every file that is already cached and which we read from the cache ... do a write to the disk!'" He further emphasized the performance impact thus:
- Atime updates are by far the biggest IO performance deficiency that Linux has today. Getting rid of atime updates would give us more everyday Linux performance than all the pagecache speedups of the past 10 years, _combined_.
Current versions of Linux, Mac OS X, Solaris, FreeBSD, NetBSD, and OpenBSD support a
Template:Visible anchor mount option, which causes the atime field never to be updated. This breaks compliance with POSIX.
Current versions of Linux support four mount options, which can be specified in fstab:
strictatime(formerly atime, and formerly the default; strictatime as of 2.6.30) – always update atime.
relatime– (relative atime), introduced in 2.6.20 and the default as of 2.6.30
nodiratime– do not update atime of directories ever
noatime– do not update atime ever; includes nodiratime; highest performance, least compatible
strictatime accords with POSIX; Alan Cox described the alternatives as:
- Turn off atime and it is very non standards compliant, turn to relatime and it is not standards compliant but nobody will break (which is good)
File systems mounted with the
noatime option do not update the atime on reads, and the
relatime option provides for updates only if the previous atime is older than the mtime or ctime, or the previous atime is over 24 hours in the past. Many users use
noatime without problem, so long as they do not use an application which depends on atime, and this offers some benefits over
relatime (no writing of atime ever on read).
As of 2.6.30 (9 June 2009), Linux defaults to
relatime, so that it will not update atime on all file reads. The behavior offers sufficient performance for most purposes and should not break any significant applications. Extended discussion of filesystem performance preceded decision. Indeed,
relatime by default was the first patch Linux applied following the 2.6.29 release. In initial patches
relatime only updated atime if atime < mtime or atime < ctime; this was subsequently modified to update atimes that were 24 hours old or older, so that tmpwatch and Debian's popularity counter (popcon) would behave properly.
ctime has nothing to do with file creation time. It is updated any time file content changes (together with
mtime), and also by changes in metadata such as file permissions, file ownership, and creation and deletion of hard links. In some implementations,
ctime is affected by renaming a file (both original Unix and modern Linux tend to do this).
ctime cannot be set with
utime() (as used e.g. by
touch); the only way to set it to an arbitrary value is by changing the system clock.
Granularity of mtime etc.Edit
time_t provides times accurate to 1 second.
Some filesystems provide greater granularity. In linux kernels 2.5.48 and above, the stat structure supports nanosecond resolution for the three file timestamp fields. These are exposed as additional fields in the stat structure.
The FAT filesystem provides timestamps with a granularity of 2 seconds.
- Kernel Trap: Linux: Replacing atime With relatime, by Jeremy, August 7, 2007
- Once upon atime, LWN, by Jonathan Corbet, August 8, 2007
- Linux 2 6 30, Linux Kernel Newbies
- That massive filesystem thread, LWN, by Jonathan Corbet, March 31, 2009
- Installing Linux on USB – Part 4: noatime and relatime mount options
- Relatime Recap, Valerie Aurora
- How accurate is ruby mtime and friends at StackOverflow.com