readlink(2)

๐Ÿ’– NAME ๐Ÿ’–

okay so this is readlink(2) from the System Calls Manual โ€” it's literally the page for readlink, readlinkat โ€” read value of a symbolic link ๐Ÿฆ‹


๐Ÿ“š LIBRARY ๐Ÿ“š

Standard C library (libc, -lc) โ€” yes bestie, the classic ๐Ÿ’…


โœจ SYNOPSIS โœจ

#include <unistd.h>

ssize_t readlink(size_t bufsiz;
                 const char *restrict path,
                 char buf[restrict bufsiz], size_t bufsiz);

#include <fcntl.h>            /* Definition of AT_* constants */
#include <unistd.h>

ssize_t readlinkat(size_t bufsiz;
                 int dirfd, const char *restrict path,
                 char buf[restrict bufsiz], size_t bufsiz);

๐Ÿงš Feature Test Macro Requirements for glibc (see feature_test_macros(7)) ๐Ÿงš

readlink():
    _XOPEN_SOURCE >= 500 || _POSIX_C_SOURCE >= 200112L
        || /* glibc <= 2.19: */ _BSD_SOURCE

readlinkat():
    Since glibc 2.10:
        _POSIX_C_SOURCE >= 200809L
    Before glibc 2.10:
        _ATFILE_SOURCE

๐ŸŒธ DESCRIPTION ๐ŸŒธ

So readlink() takes the contents of the symbolic link path and places them in the buffer buf, which has size bufsiz. And here's the thing nobody tells you: readlink() does not append a terminating null byte to buf. Like, at all. Not even a little one. ๐Ÿซ 

It will (silently) truncate the contents โ€” down to a length of bufsiz characters โ€” in case the buffer is too small to hold all of the contents. Silently!! no warning, no drama, just vibes and lost data ๐Ÿ˜ญ

๐Ÿ”ฅ readlinkat() ๐Ÿ”ฅ

The readlinkat() system call operates in exactly the same way as readlink(), except for the differences described here. It's giving "same energy, tiny plot twist".

  • If path is relative, then it is interpreted relative to the directory referred to by the file descriptor dirfd (rather than relative to the current working directory of the calling process, as is done by readlink() for a relative pathname).
  • If path is relative and dirfd is the special value AT_FDCWD, then path is interpreted relative to the current working directory of the calling process (like readlink()).
  • If path is absolute, then dirfd is ignored.

Since Linux 2.6.39, path can be an empty string, in which case the call operates on the symbolic link referred to by dirfd (which should have been obtained using open(2) with the O_PATH and O_NOFOLLOW flags). yes this is real, empty string, main character behavior ๐Ÿ‘‘

See openat(2) for an explanation of the need for readlinkat().


๐Ÿ’ซ RETURN VALUE ๐Ÿ’ซ

On success, these calls return the number of bytes placed in buf. (If the returned value equals bufsiz, then truncation may have occurred โ€” ๐Ÿ‘€ keep an eye on that.)

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


๐Ÿ’€ ERRORS ๐Ÿ’€

  • EACCES โ€” Search permission is denied for a component of the path prefix. (See also path_resolution(7).)
  • EBADF โ€” (readlinkat()) path is relative but dirfd is neither AT_FDCWD nor a valid file descriptor.
  • EFAULT โ€” buf extends outside the process's allocated address space. bestie your pointer is somewhere it should NOT be ๐Ÿ™ˆ
  • EINVAL โ€” bufsiz is not positive.
  • EINVAL โ€” The named file (i.e., the final filename component of path) is not a symbolic link.
  • EIO โ€” An I/O error occurred while reading from the filesystem.
  • ELOOP โ€” Too many symbolic links were encountered in translating the pathname. it's links all the way down ๐Ÿ’€
  • ENAMETOOLONG โ€” A pathname, or a component of a pathname, was too long.
  • ENOENT โ€” The named file does not exist.
  • ENOMEM โ€” Insufficient kernel memory was available.
  • ENOTDIR โ€” A component of the path prefix is not a directory.
  • ENOTDIR โ€” (readlinkat()) path is relative and dirfd is a file descriptor referring to a file other than a directory.

โญ STANDARDS โญ

POSIX.1-2024. periodt.


๐Ÿฆ‹ HISTORY ๐Ÿฆ‹

  • readlink() โ€” 4.4BSD (first appeared in 4.2BSD), POSIX.1-2001, POSIX.1-2008.
  • readlinkat() โ€” POSIX.1-2008. Linux 2.6.16, glibc 2.4.

Up to and including glibc 2.4, the return type of readlink() was declared as int. Nowadays, the return type is declared as ssize_t, as (newly) required in POSIX.1-2001. glow-up fr fr โœจ

๐ŸŽ€ glibc ๐ŸŽ€

On older kernels where readlinkat() is unavailable, the glibc wrapper function falls back to the use of readlink(). When path is relative, glibc constructs a pathname based on the symbolic link in /proc/self/fd that corresponds to the dirfd argument. sneaky little workaround, iconic ๐Ÿ’…


๐Ÿง  NOTES ๐Ÿง 

Using a statically sized buffer might not provide enough room for the symbolic link contents. The required size for the buffer can be obtained from the stat.st_size value returned by a call to lstat(2) on the link.

However โ€” and this is the important part โ€” the number of bytes written by readlink() and readlinkat() should be checked to make sure that the size of the symbolic link did not increase between the calls. race condition girlies rise ๐Ÿ™ˆ

Dynamically allocating the buffer for readlink() and readlinkat() also addresses a common portability problem when using PATH_MAX for the buffer size, as this constant is not guaranteed to be defined per POSIX if the system does not have such limit.


๐ŸŒˆ EXAMPLES ๐ŸŒˆ

The following program allocates the buffer needed by readlink() dynamically from the information provided by lstat(2), falling back to a buffer of size PATH_MAX in cases where lstat(2) reports a size of zero.

#include <limits.h>
#include <stdio.h>
#include <stdlib.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <unistd.h>

int
main(int argc, char *argv[])
{
    char         *buf;
    ssize_t      nbytes, bufsiz;
    struct stat  sb;

    if (argc != 2) {
        fprintf(stderr, "Usage: %s <path>\n", argv[0]);
        exit(EXIT_FAILURE);
    }

    if (lstat(argv[1], &sb) == -1) {
        perror("lstat");
        exit(EXIT_FAILURE);
    }

    /* Add one to the link size, so that we can determine whether
       the buffer returned by readlink() was truncated.  */

    bufsiz = sb.st_size + 1;

    /* Some magic symlinks under (for example) /proc and /sys
       report 'st_size' as zero.  In that case, take PATH_MAX as
       a "good enough" estimate.  */

    if (sb.st_size == 0)
        bufsiz = PATH_MAX;

    buf = malloc(bufsiz);
    if (buf == NULL) {
        perror("malloc");
        exit(EXIT_FAILURE);
    }

    nbytes = readlink(argv[1], buf, bufsiz);
    if (nbytes == -1) {
        perror("readlink");
        exit(EXIT_FAILURE);
    }

    /* Print only 'nbytes' of 'buf', as it doesn't contain a terminating
       null byte ('\0').  */
    printf("'%s' points to '%.*s'\n", argv[1], (int) nbytes, buf);

    /* If the return value was equal to the buffer size, then
       the link target was larger than expected (perhaps because the
       target was changed between the call to lstat() and the call to
       readlink()).  Warn the user that the returned target may have
       been truncated.  */

    if (nbytes == bufsiz)
        printf("(Returned buffer may have been truncated)\n");

    free(buf);
    exit(EXIT_SUCCESS);
}

๐Ÿ‘‘ SEE ALSO ๐Ÿ‘‘

readlink(1), lstat(2), stat(2), symlink(2), realpath(3), path_resolution(7), symlink(7)


okay that's the whole page bestie, go read some symlinks ๐Ÿ’– this lil love letter is from Linux man-pages 6.20, dated 2026-02-08 โ€” she's fresh, she's relevant, she's readlink(2) ๐ŸŽ€โœจ

๐Ÿ“ฆ where this page comes from
  • arch rolling: man-pages 6.20-1 ยท BSD-2-Clause AND BSD-3-Clause AND BSD-4.3TAHOE AND BSD-4-Clause-UC AND GPL-1.0-or-later AND GPL-2.0-only AND GPL-2.0-or-later AND GPL-3.0-or-later AND LGPL-3.0-linking-exception AND LGPL-3.0-or-later AND LicenseRef-Public-Domain AND LicenseRef-UltraPermissive AND Linux-man-pages-1-para AND Linux-man-pages-copyleft AND Linux-man-pages-copyleft-2-para AND Linux-man-pages-copyleft-var AND MIT AND Spencer-94
  • nixpkgs unstable: man-pages 6.20
  • ๐Ÿ“œ the page's license: BSD-4-Clause-UC
  • ๐Ÿ’– girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms