Scroll to navigation

statvfs(3) Library Functions Manual statvfs(3)


statvfs, fstatvfs - get filesystem statistics


Standard C library (libc, -lc)


#include <sys/statvfs.h>
int statvfs(const char *restrict path, struct statvfs *restrict buf);
int fstatvfs(int fd, struct statvfs *buf);


The function statvfs() returns information about a mounted filesystem. path is the pathname of any file within the mounted filesystem. buf is a pointer to a statvfs structure defined approximately as follows:

struct statvfs {

unsigned long f_bsize; /* Filesystem block size */
unsigned long f_frsize; /* Fragment size */
fsblkcnt_t f_blocks; /* Size of fs in f_frsize units */
fsblkcnt_t f_bfree; /* Number of free blocks */
fsblkcnt_t f_bavail; /* Number of free blocks for
unprivileged users */
fsfilcnt_t f_files; /* Number of inodes */
fsfilcnt_t f_ffree; /* Number of free inodes */
fsfilcnt_t f_favail; /* Number of free inodes for
unprivileged users */
unsigned long f_fsid; /* Filesystem ID */
unsigned long f_flag; /* Mount flags */
unsigned long f_namemax; /* Maximum filename length */ };

Here the types fsblkcnt_t and fsfilcnt_t are defined in <sys/types.h>. Both used to be unsigned long.

The field f_flag is a bit mask indicating various options that were employed when mounting this filesystem. It contains zero or more of the following flags:

Mandatory locking is permitted on the filesystem (see fcntl(2)).
Do not update access times; see mount(2).
Disallow access to device special files on this filesystem.
Do not update directory access times; see mount(2).
Execution of programs is disallowed on this filesystem.
The set-user-ID and set-group-ID bits are ignored by exec(3) for executable files on this filesystem
This filesystem is mounted read-only.
Update atime relative to mtime/ctime; see mount(2).
Writes are synched to the filesystem immediately (see the description of O_SYNC in open(2)).

It is unspecified whether all members of the returned struct have meaningful values on all filesystems.

fstatvfs() returns the same information about an open file referenced by descriptor fd.


On success, zero is returned. On error, -1 is returned, and errno is set to indicate the error.


(statvfs()) Search permission is denied for a component of the path prefix of path. (See also path_resolution(7).)
(fstatvfs()) fd is not a valid open file descriptor.
Buf or path points to an invalid address.
This call was interrupted by a signal; see signal(7).
An I/O error occurred while reading from the filesystem.
(statvfs()) Too many symbolic links were encountered in translating path.
(statvfs()) path is too long.
(statvfs()) The file referred to by path does not exist.
Insufficient kernel memory was available.
The filesystem does not support this call.
(statvfs()) A component of the path prefix of path is not a directory.
Some values were too large to be represented in the returned struct.


For an explanation of the terms used in this section, see attributes(7).

Interface Attribute Value
statvfs (), fstatvfs () Thread safety MT-Safe


Only the ST_NOSUID and ST_RDONLY flags of the f_flag field are specified in POSIX.1. To obtain definitions of the remaining flags, one must define _GNU_SOURCE.


The Linux kernel has system calls statfs(2) and fstatfs(2) to support this library call.

The glibc implementations of

pathconf(path, _PC_REC_XFER_ALIGN);
pathconf(path, _PC_ALLOC_SIZE_MIN);
pathconf(path, _PC_REC_MIN_XFER_SIZE);

respectively use the f_frsize, f_frsize, and f_bsize fields returned by a call to statvfs() with the argument path.

Under Linux, f_favail is always the same as f_ffree, and there's no way for a filesystem to report otherwise. This is not an issue, since no filesystems with an inode root reservation exist.





Before glibc 2.13, statvfs() populated the bits of the f_flag field by scanning the mount options shown in /proc/mounts. However, starting with Linux 2.6.36, the underlying statfs(2) system call provides the necessary information via the f_flags field, and since glibc 2.13, the statvfs() function will use information from that field rather than scanning /proc/mounts.



2024-05-02 Linux man-pages (unreleased)