Hi,

As per my man pages, the stat structure contains the following fields:
struct stat {
dev_t         st_dev;      /* device */
ino_t         st_ino;      /* inode */
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 type (if inode
device) */
off_t         st_size;     /* total size, in bytes
*/
blksize_t     st_blksize;  /* blocksize for
filesystem I/O */
blkcnt_t      st_blocks;   /* number of blocks
allocated */
time_t        st_atime;    /* time of last access */
time_t        st_mtime;    /* time of last
modification */
time_t        st_ctime;    /* time of last change */
};

I am interested in the file creation time.  What field should I use?
Should I go with some other system call?  Also, what is the difference
between the time of last modification and the time of last change?

Thanks,
Gus

[ Post a follow-up to this message ]

On Dec 28, 4:21 pm, Generic Usenet Account <use...@sta.samsung.com>
wrote:
> Hi,
>
> As per my man pages, the stat structure contains the following fields:
>               struct stat {
>                   dev_t         st_dev;      /* device */
>                   ino_t         st_ino;      /* inode */
>                   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 type (if inode
> device) */
>                   off_t         st_size;     /* total size, in bytes
> */
>                   blksize_t     st_blksize;  /* blocksize for
> filesystem I/O */
>                   blkcnt_t      st_blocks;   /* number of blocks
> allocated */
>                   time_t        st_atime;    /* time of last access */
>                   time_t        st_mtime;    /* time of last
> modification */
>                   time_t        st_ctime;    /* time of last change */
>               };
>
> I am interested in the file creation time.  What field should I use?

None of these tell you the file creation time.  That information is
only kept for a few filesystems.  For example, ufs2 on FreeBSD fills
in an st_birthtime field.  But in general it's not available via
stat(2) or any other system call.

> Should I go with some other system call?  Also, what is the difference
> between the time of last modification and the time of last change?

The FreeBSD man page describes them as follows:

st_atime      Time when file data last accessed.  Changed by the
mknod(2), utimes(2), read(2) and readv(2) system
calls.

st_mtime      Time when file data last modified.  Changed by the
mkdir(2), mkfifo(2), mknod(2), utimes(2), write(2)
and
writev(2) system calls.

st_ctime      Time when file status was last changed (inode data
modifi-
cation).  Changed by the chflags(2), chmod(2),
chown(2),
creat(2), link(2), mkdir(2), mkfifo(2), mknod(2),
rename(2), rmdir(2), symlink(2), truncate(2),
unlink(2),
utimes(2), write(2) and writev(2) system calls.

st_birthtime  Time when the inode was created.

Hope this helps.

[ Post a follow-up to this message ]

>              struct stat {
...
>                  time_t        st_atime;    /* time of last access */
>                  time_t        st_mtime;    /* time of last
>modification */
>                  time_t        st_ctime;    /* time of last change */
>
>I am interested in the file creation time.

There is no file creation time here.  I don't even think there is a
coherent definition of "file creation time".  For example, should the
shell command:
echo foo > file
update the so-called "creation time" if file "file" existed before
running this command?

>What field should I use?

st_mtime is probably pretty close for files which are written once
and not modified later.  For write-frequently database files, you're
out of luck.

>Should I go with some other system call?

What makes you think a file creation time is stored anywhere?

>Also, what is the difference
>between the time of last modification and the time of last change?

Operations such as chown and chmod change st_ctime but not st_mtime.

[ Post a follow-up to this message ]