1statfs(2) System Calls Manual statfs(2)
2
6 statfs, fstatfs - get filesystem statistics
7
9 Standard C library (libc, -lc)
10
12 #include <sys/vfs.h> /* or <sys/statfs.h> */
13 int statfs(const char *path, struct statfs *buf);
15 int fstatfs(int fd, struct statfs *buf);
16 Unless you need the f_type field, you should use the standard
18 statvfs(3) interface instead.
19
21 The statfs() system call returns information about a mounted filesys‐
22 tem. path is the pathname of any file within the mounted filesystem.
23 buf is a pointer to a statfs structure defined approximately as fol‐
24 lows:
25 struct statfs {
27 __fsword_t f_type; /* Type of filesystem (see below) */
28 __fsword_t f_bsize; /* Optimal transfer block size */
29 fsblkcnt_t f_blocks; /* Total data blocks in filesystem */
30 fsblkcnt_t f_bfree; /* Free blocks in filesystem */
31 fsblkcnt_t f_bavail; /* Free blocks available to
32 unprivileged user */
33 fsfilcnt_t f_files; /* Total inodes in filesystem */
34 fsfilcnt_t f_ffree; /* Free inodes in filesystem */
35 fsid_t f_fsid; /* Filesystem ID */
36 __fsword_t f_namelen; /* Maximum length of filenames */
37 __fsword_t f_frsize; /* Fragment size (since Linux 2.6) */
38 __fsword_t f_flags; /* Mount flags of filesystem
39 (since Linux 2.6.36) */
40 __fsword_t f_spare[xxx];
41 /* Padding bytes reserved for future use */
42 };
43 The following filesystem types may appear in f_type:
45 ADFS_SUPER_MAGIC 0xadf5
47 AFFS_SUPER_MAGIC 0xadff
48 AFS_SUPER_MAGIC 0x5346414f
49 ANON_INODE_FS_MAGIC 0x09041934 /* Anonymous inode FS (for
50 pseudofiles that have no name;
51 e.g., epoll, signalfd, bpf) */
52 AUTOFS_SUPER_MAGIC 0x0187
53 BDEVFS_MAGIC 0x62646576
54 BEFS_SUPER_MAGIC 0x42465331
55 BFS_MAGIC 0x1badface
56 BINFMTFS_MAGIC 0x42494e4d
57 BPF_FS_MAGIC 0xcafe4a11
58 BTRFS_SUPER_MAGIC 0x9123683e
59 BTRFS_TEST_MAGIC 0x73727279
60 CGROUP_SUPER_MAGIC 0x27e0eb /* Cgroup pseudo FS */
61 CGROUP2_SUPER_MAGIC 0x63677270 /* Cgroup v2 pseudo FS */
62 CIFS_MAGIC_NUMBER 0xff534d42
63 CODA_SUPER_MAGIC 0x73757245
64 COH_SUPER_MAGIC 0x012ff7b7
65 CRAMFS_MAGIC 0x28cd3d45
66 DEBUGFS_MAGIC 0x64626720
67 DEVFS_SUPER_MAGIC 0x1373 /* Linux 2.6.17 and earlier */
68 DEVPTS_SUPER_MAGIC 0x1cd1
69 ECRYPTFS_SUPER_MAGIC 0xf15f
70 EFIVARFS_MAGIC 0xde5e81e4
71 EFS_SUPER_MAGIC 0x00414a53
72 EXT_SUPER_MAGIC 0x137d /* Linux 2.0 and earlier */
73 EXT2_OLD_SUPER_MAGIC 0xef51
74 EXT2_SUPER_MAGIC 0xef53
75 EXT3_SUPER_MAGIC 0xef53
76 EXT4_SUPER_MAGIC 0xef53
77 F2FS_SUPER_MAGIC 0xf2f52010
78 FUSE_SUPER_MAGIC 0x65735546
79 FUTEXFS_SUPER_MAGIC 0xbad1dea /* Unused */
80 HFS_SUPER_MAGIC 0x4244
81 HOSTFS_SUPER_MAGIC 0x00c0ffee
82 HPFS_SUPER_MAGIC 0xf995e849
83 HUGETLBFS_MAGIC 0x958458f6
84 ISOFS_SUPER_MAGIC 0x9660
85 JFFS2_SUPER_MAGIC 0x72b6
86 JFS_SUPER_MAGIC 0x3153464a
87 MINIX_SUPER_MAGIC 0x137f /* original minix FS */
88 MINIX_SUPER_MAGIC2 0x138f /* 30 char minix FS */
89 MINIX2_SUPER_MAGIC 0x2468 /* minix V2 FS */
90 MINIX2_SUPER_MAGIC2 0x2478 /* minix V2 FS, 30 char names */
91 MINIX3_SUPER_MAGIC 0x4d5a /* minix V3 FS, 60 char names */
92 MQUEUE_MAGIC 0x19800202 /* POSIX message queue FS */
93 MSDOS_SUPER_MAGIC 0x4d44
94 MTD_INODE_FS_MAGIC 0x11307854
95 NCP_SUPER_MAGIC 0x564c
96 NFS_SUPER_MAGIC 0x6969
97 NILFS_SUPER_MAGIC 0x3434
98 NSFS_MAGIC 0x6e736673
99 NTFS_SB_MAGIC 0x5346544e
100 OCFS2_SUPER_MAGIC 0x7461636f
101 OPENPROM_SUPER_MAGIC 0x9fa1
102 OVERLAYFS_SUPER_MAGIC 0x794c7630
103 PIPEFS_MAGIC 0x50495045
104 PROC_SUPER_MAGIC 0x9fa0 /* /proc FS */
105 PSTOREFS_MAGIC 0x6165676c
106 QNX4_SUPER_MAGIC 0x002f
107 QNX6_SUPER_MAGIC 0x68191122
108 RAMFS_MAGIC 0x858458f6
109 REISERFS_SUPER_MAGIC 0x52654973
110 ROMFS_MAGIC 0x7275
111 SECURITYFS_MAGIC 0x73636673
112 SELINUX_MAGIC 0xf97cff8c
113 SMACK_MAGIC 0x43415d53
114 SMB_SUPER_MAGIC 0x517b
115 SMB2_MAGIC_NUMBER 0xfe534d42
116 SOCKFS_MAGIC 0x534f434b
117 SQUASHFS_MAGIC 0x73717368
118 SYSFS_MAGIC 0x62656572
119 SYSV2_SUPER_MAGIC 0x012ff7b6
120 SYSV4_SUPER_MAGIC 0x012ff7b5
121 TMPFS_MAGIC 0x01021994
122 TRACEFS_MAGIC 0x74726163
123 UDF_SUPER_MAGIC 0x15013346
124 UFS_MAGIC 0x00011954
125 USBDEVICE_SUPER_MAGIC 0x9fa2
126 V9FS_MAGIC 0x01021997
127 VXFS_SUPER_MAGIC 0xa501fcf5
128 XENFS_SUPER_MAGIC 0xabba1974
129 XENIX_SUPER_MAGIC 0x012ff7b4
130 XFS_SUPER_MAGIC 0x58465342
131 _XIAFS_SUPER_MAGIC 0x012fd16d /* Linux 2.0 and earlier */
132 Most of these MAGIC constants are defined in /usr/in‐
134 clude/linux/magic.h, and some are hardcoded in kernel sources.
135 The f_flags field is a bit mask indicating mount options for the
137 filesystem. It contains zero or more of the following bits:
138 ST_MANDLOCK
140 Mandatory locking is permitted on the filesystem (see fcntl(2)).
141 ST_NOATIME
143 Do not update access times; see mount(2).
144 ST_NODEV
146 Disallow access to device special files on this filesystem.
147 ST_NODIRATIME
149 Do not update directory access times; see mount(2).
150 ST_NOEXEC
152 Execution of programs is disallowed on this filesystem.
153 ST_NOSUID
155 The set-user-ID and set-group-ID bits are ignored by exec(3) for
156 executable files on this filesystem
157 ST_RDONLY
159 This filesystem is mounted read-only.
160 ST_RELATIME
162 Update atime relative to mtime/ctime; see mount(2).
163 ST_SYNCHRONOUS
165 Writes are synched to the filesystem immediately (see the de‐
166 scription of O_SYNC in open(2)).
167 ST_NOSYMFOLLOW (since Linux 5.10)
169 Symbolic links are not followed when resolving paths; see
170 mount(2).
171 Nobody knows what f_fsid is supposed to contain (but see below).
173 Fields that are undefined for a particular filesystem are set to 0.
175 fstatfs() returns the same information about an open file referenced by
177 descriptor fd.
178
180 On success, zero is returned. On error, -1 is returned, and errno is
181 set to indicate the error.
182
184 EACCES (statfs()) Search permission is denied for a component of the
185 path prefix of path. (See also path_resolution(7).)
186 EBADF (fstatfs()) fd is not a valid open file descriptor.
188 EFAULT buf or path points to an invalid address.
190 EINTR The call was interrupted by a signal; see signal(7).
192 EIO An I/O error occurred while reading from the filesystem.
194 ELOOP (statfs()) Too many symbolic links were encountered in translat‐
196 ing path.
197 ENAMETOOLONG
199 (statfs()) path is too long.
200 ENOENT (statfs()) The file referred to by path does not exist.
202 ENOMEM Insufficient kernel memory was available.
204 ENOSYS The filesystem does not support this call.
206 ENOTDIR
208 (statfs()) A component of the path prefix of path is not a di‐
209 rectory.
210 EOVERFLOW
212 Some values were too large to be represented in the returned
213 struct.
214
216 The f_fsid field
217 Solaris, Irix, and POSIX have a system call statvfs(2) that returns a
218 struct statvfs (defined in <sys/statvfs.h>) containing an unsigned long
219 f_fsid. Linux, SunOS, HP-UX, 4.4BSD have a system call statfs() that
220 returns a struct statfs (defined in <sys/vfs.h>) containing a fsid_t
221 f_fsid, where fsid_t is defined as struct { int val[2]; }. The same
222 holds for FreeBSD, except that it uses the include file <sys/mount.h>.
223 The general idea is that f_fsid contains some random stuff such that
225 the pair (f_fsid,ino) uniquely determines a file. Some operating sys‐
226 tems use (a variation on) the device number, or the device number com‐
227 bined with the filesystem type. Several operating systems restrict
228 giving out the f_fsid field to the superuser only (and zero it for un‐
229 privileged users), because this field is used in the filehandle of the
230 filesystem when NFS-exported, and giving it out is a security concern.
231 Under some operating systems, the fsid can be used as the second argu‐
233 ment to the sysfs(2) system call.
234
236 Linux.
237
239 The Linux statfs() was inspired by the 4.4BSD one (but they do not use
240 the same structure).
241 The original Linux statfs() and fstatfs() system calls were not de‐
243 signed with extremely large file sizes in mind. Subsequently, Linux
244 2.6 added new statfs64() and fstatfs64() system calls that employ a new
245 structure, statfs64. The new structure contains the same fields as the
246 original statfs structure, but the sizes of various fields are in‐
247 creased, to accommodate large file sizes. The glibc statfs() and fs‐
248 tatfs() wrapper functions transparently deal with the kernel differ‐
249 ences.
250 LSB has deprecated the library calls statfs() and fstatfs() and tells
252 us to use statvfs(3) and fstatvfs(3) instead.
253
255 The __fsword_t type used for various fields in the statfs structure
256 definition is a glibc internal type, not intended for public use. This
257 leaves the programmer in a bit of a conundrum when trying to copy or
258 compare these fields to local variables in a program. Using un‐
259 signed int for such variables suffices on most systems.
260 Some systems have only <sys/vfs.h>, other systems also have
262 <sys/statfs.h>, where the former includes the latter. So it seems in‐
263 cluding the former is the best choice.
264
266 From Linux 2.6.38 up to and including Linux 3.1, fstatfs() failed with
267 the error ENOSYS for file descriptors created by pipe(2).
268
270 stat(2), statvfs(3), path_resolution(7)
271Linux man-pages 6.05 2023-07-18 statfs(2)