1 .\" This manpage is Copyright (C) 1992 Drew Eckhardt;
2 .\" and Copyright (C) 1993 Michael Haardt
3 .\" and Copyright (C) 1993,1994 Ian Jackson
4 .\" and Copyright (C) 2006, 2014 Michael Kerrisk
6 .\" %%%LICENSE_START(GPL_NOVERSION_ONELINE)
7 .\" You may distribute it under the terms of the GNU General
8 .\" Public License. It comes with NO WARRANTY.
11 .TH MKDIR 2 2017-05-03 "Linux" "Linux Programmer's Manual"
13 mkdir, mkdirat \- create a directory
16 .B #include <sys/stat.h>
17 .B #include <sys/types.h>
18 .\" .B #include <unistd.h>
20 .BI "int mkdir(const char *" pathname ", mode_t " mode );
22 .BR "#include <fcntl.h> " "/* Definition of AT_* constants */"
23 .B #include <sys/stat.h>
25 .BI "int mkdirat(int " dirfd ", const char *" pathname ", mode_t " mode );
29 Feature Test Macro Requirements for glibc (see
30 .BR feature_test_macros (7)):
39 _POSIX_C_SOURCE\ >=\ 200809L
49 attempts to create a directory named
54 specifies the mode for the new directory (see
56 It is modified by the process's
58 in the usual way: in the absence of a default ACL, the mode of the
60 .RI ( mode " & ~" umask " & 0777)."
63 bits are honored for the created directory depends on the operating system.
64 For Linux, see NOTES below.
66 The newly created directory will be owned by the effective user ID of the
68 If the directory containing the file has the set-group-ID
69 bit set, or if the filesystem is mounted with BSD group semantics
70 .RI ( "mount -o bsdgroups"
72 .IR "mount -o grpid" ),
73 the new directory will inherit the group ownership from its parent;
74 otherwise it will be owned by the effective group ID of the process.
76 If the parent directory has the set-group-ID bit set, then so will the
77 newly created directory.
83 system call operates in exactly the same way as
85 except for the differences described here.
87 If the pathname given in
89 is relative, then it is interpreted relative to the directory
90 referred to by the file descriptor
92 (rather than relative to the current working directory of
93 the calling process, as is done by
95 for a relative pathname).
105 is interpreted relative to the current working
106 directory of the calling process (like
117 for an explanation of the need for
123 return zero on success, or \-1 if an error occurred (in which case,
125 is set appropriately).
129 The parent directory does not allow write permission to the process,
130 or one of the directories in
132 did not allow search permission.
134 .BR path_resolution (7).)
137 The user's quota of disk blocks or inodes on the filesystem has been
142 already exists (not necessarily as a directory).
143 This includes the case where
145 is a symbolic link, dangling or not.
148 .IR pathname " points outside your accessible address space."
151 Too many symbolic links were encountered in resolving
155 The number of links to the parent directory would exceed
159 .IR pathname " was too long."
162 A directory component in
164 does not exist or is a dangling symbolic link.
167 Insufficient kernel memory was available.
170 The device containing
172 has no room for the new directory.
175 The new directory cannot be created because the user's disk quota is
179 A component used as a directory in
181 is not, in fact, a directory.
184 The filesystem containing
186 does not support the creation of directories.
190 refers to a file on a read-only filesystem.
192 The following additional errors can occur for
197 is not a valid file descriptor.
203 is a file descriptor referring to a file other than a directory.
206 was added to Linux in kernel 2.6.16;
207 library support was added to glibc in version 2.4.
210 SVr4, BSD, POSIX.1-2001, POSIX.1-2008.
211 .\" SVr4 documents additional EIO, EMULTIHOP
216 Under Linux, apart from the permission bits, the
221 There are many infelicities in the protocol underlying NFS.
225 On older kernels where
227 is unavailable, the glibc wrapper function falls back to the use of
231 is a relative pathname,
232 glibc constructs a pathname based on the symbolic link in
234 that corresponds to the
248 .BR path_resolution (7)