man/man5/proc_pid_root.5

.\" Copyright (C) 1994, 1995, Daniel Quinlan <quinlan@yggdrasil.com>
.\" Copyright (C) 2002-2008, 2017, Michael Kerrisk <mtk.manpages@gmail.com>
.\" Copyright (C) 2023, Alejandro Colomar <alx@kernel.org>
.\"
.\" SPDX-License-Identifier: GPL-3.0-or-later
.\"
.TH proc_pid_root 5 (date) "Linux man-pages (unreleased)"
.SH NAME
/proc/pid/root/ \- symbolic link to root directory
.SH DESCRIPTION
.TP
.IR /proc/ pid /root/
UNIX and Linux support the idea of a per-process root of the
filesystem, set by the
.BR chroot (2)
system call.
This file is a symbolic link that points to the process's
root directory, and behaves in the same way as
.IR exe ,
and
.IR fd/* .
.IP
Note however that this file is not merely a symbolic link.
It provides the same view of the filesystem (including namespaces and the
set of per-process mounts) as the process itself.
An example illustrates this point.
In one terminal, we start a shell in new user and mount namespaces,
and in that shell we create some new mounts:
.IP
.in +4n
.EX
$ \fBPS1=\[aq]sh1# \[aq] unshare \-Urnm\fP
sh1# \fBmount \-t tmpfs tmpfs /etc\fP  # Mount empty tmpfs at /etc
sh1# \fBmount \-\-bind /usr /dev\fP     # Mount /usr at /dev
sh1# \fBecho $$\fP
27123
.EE
.in
.IP
In a second terminal window, in the initial mount namespace,
we look at the contents of the corresponding mounts in
the initial and new namespaces:
.IP
.in +4n
.EX
$ \fBPS1=\[aq]sh2# \[aq] sudo sh\fP
sh2# \fBls /etc | wc \-l\fP                  # In initial NS
309
sh2# \fBls /proc/27123/root/etc | wc \-l\fP  # /etc in other NS
0                                     # The empty tmpfs dir
sh2# \fBls /dev | wc \-l\fP                  # In initial NS
205
sh2# \fBls /proc/27123/root/dev | wc \-l\fP  # /dev in other NS
11                                    # Actually bind
                                      # mounted to /usr
sh2# \fBls /usr | wc \-l\fP                  # /usr in initial NS
11
.EE
.in
.IP
.\" The following was still true as at kernel 2.6.13
In a multithreaded process, the contents of the
.IR /proc/ pid /root
symbolic link are not available if the main thread has already terminated
(typically by calling
.BR pthread_exit (3)).
.IP
Permission to dereference or read
.RB ( readlink (2))
this symbolic link is governed by a ptrace access mode
.B PTRACE_MODE_READ_FSCREDS
check; see
.BR ptrace (2).
.SH SEE ALSO
.BR proc (5)