2016-09-17 04:44:50 +08:00
|
|
|
/* Check posix_spawn add file actions.
|
2023-01-07 05:08:04 +08:00
|
|
|
Copyright (C) 2016-2023 Free Software Foundation, Inc.
|
2016-09-17 04:44:50 +08:00
|
|
|
This file is part of the GNU C Library.
|
|
|
|
|
|
|
|
The GNU C Library is free software; you can redistribute it and/or
|
|
|
|
modify it under the terms of the GNU Lesser General Public
|
|
|
|
License as published by the Free Software Foundation; either
|
|
|
|
version 2.1 of the License, or (at your option) any later version.
|
|
|
|
|
|
|
|
The GNU C Library is distributed in the hope that it will be useful,
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
Lesser General Public License for more details.
|
|
|
|
|
|
|
|
You should have received a copy of the GNU Lesser General Public
|
|
|
|
License along with the GNU C Library; if not, see
|
Prefer https to http for gnu.org and fsf.org URLs
Also, change sources.redhat.com to sourceware.org.
This patch was automatically generated by running the following shell
script, which uses GNU sed, and which avoids modifying files imported
from upstream:
sed -ri '
s,(http|ftp)(://(.*\.)?(gnu|fsf|sourceware)\.org($|[^.]|\.[^a-z])),https\2,g
s,(http|ftp)(://(.*\.)?)sources\.redhat\.com($|[^.]|\.[^a-z]),https\2sourceware.org\4,g
' \
$(find $(git ls-files) -prune -type f \
! -name '*.po' \
! -name 'ChangeLog*' \
! -path COPYING ! -path COPYING.LIB \
! -path manual/fdl-1.3.texi ! -path manual/lgpl-2.1.texi \
! -path manual/texinfo.tex ! -path scripts/config.guess \
! -path scripts/config.sub ! -path scripts/install-sh \
! -path scripts/mkinstalldirs ! -path scripts/move-if-change \
! -path INSTALL ! -path locale/programs/charmap-kw.h \
! -path po/libc.pot ! -path sysdeps/gnu/errlist.c \
! '(' -name configure \
-execdir test -f configure.ac -o -f configure.in ';' ')' \
! '(' -name preconfigure \
-execdir test -f preconfigure.ac ';' ')' \
-print)
and then by running 'make dist-prepare' to regenerate files built
from the altered files, and then executing the following to cleanup:
chmod a+x sysdeps/unix/sysv/linux/riscv/configure
# Omit irrelevant whitespace and comment-only changes,
# perhaps from a slightly-different Autoconf version.
git checkout -f \
sysdeps/csky/configure \
sysdeps/hppa/configure \
sysdeps/riscv/configure \
sysdeps/unix/sysv/linux/csky/configure
# Omit changes that caused a pre-commit check to fail like this:
# remote: *** error: sysdeps/powerpc/powerpc64/ppc-mcount.S: trailing lines
git checkout -f \
sysdeps/powerpc/powerpc64/ppc-mcount.S \
sysdeps/unix/sysv/linux/s390/s390-64/syscall.S
# Omit change that caused a pre-commit check to fail like this:
# remote: *** error: sysdeps/sparc/sparc64/multiarch/memcpy-ultra3.S: last line does not end in newline
git checkout -f sysdeps/sparc/sparc64/multiarch/memcpy-ultra3.S
2019-09-07 13:40:42 +08:00
|
|
|
<https://www.gnu.org/licenses/>. */
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <spawn.h>
|
|
|
|
#include <error.h>
|
|
|
|
#include <errno.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <unistd.h>
|
|
|
|
#include <sys/wait.h>
|
|
|
|
#include <sys/resource.h>
|
2017-04-25 02:48:01 +08:00
|
|
|
#include <fcntl.h>
|
|
|
|
#include <paths.h>
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
#include <intprops.h>
|
2016-09-17 04:44:50 +08:00
|
|
|
|
2017-04-25 02:48:01 +08:00
|
|
|
#include <support/check.h>
|
|
|
|
#include <support/temp_file.h>
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
#include <support/xunistd.h>
|
|
|
|
#include <tst-spawn.h>
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
static int
|
|
|
|
do_test (void)
|
|
|
|
{
|
|
|
|
/* The test checks if posix_spawn open file action close the file descriptor
|
|
|
|
before opening a new one in case the input file descriptor is already
|
|
|
|
opened. It does by exhausting all file descriptors on the process before
|
|
|
|
issue posix_spawn. It then issues a posix_spawn for '/bin/sh echo $$'
|
|
|
|
and add two rules:
|
|
|
|
|
|
|
|
1. Redirect stdout to a temporary filepath
|
|
|
|
2. Redirect stderr to stdout
|
|
|
|
|
|
|
|
If the implementation does not close the file 1. will fail with
|
|
|
|
EMFILE. */
|
|
|
|
|
|
|
|
struct rlimit rl;
|
|
|
|
int max_fd = 24;
|
|
|
|
|
|
|
|
/* Set maximum number of file descriptor to a low value to avoid open
|
|
|
|
too many files in environments where RLIMIT_NOFILE is large and to
|
|
|
|
limit the array size to track the opened file descriptors. */
|
|
|
|
|
|
|
|
if (getrlimit (RLIMIT_NOFILE, &rl) == -1)
|
2017-04-25 02:48:01 +08:00
|
|
|
FAIL_EXIT1 ("getrlimit (RLIMIT_NOFILE): %m");
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
max_fd = (rl.rlim_cur < max_fd ? rl.rlim_cur : max_fd);
|
|
|
|
rl.rlim_cur = max_fd;
|
|
|
|
|
|
|
|
if (setrlimit (RLIMIT_NOFILE, &rl) == 1)
|
2017-04-25 02:48:01 +08:00
|
|
|
FAIL_EXIT1 ("setrlimit (RLIMIT_NOFILE): %m");
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
/* Exhauste the file descriptor limit with temporary files. */
|
|
|
|
int files[max_fd];
|
|
|
|
int nfiles = 0;
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
for (; nfiles < max_fd; nfiles++)
|
2016-09-17 04:44:50 +08:00
|
|
|
{
|
|
|
|
int fd = create_temp_file ("tst-spawn3.", NULL);
|
|
|
|
if (fd == -1)
|
|
|
|
{
|
|
|
|
if (errno != EMFILE)
|
2017-04-25 02:48:01 +08:00
|
|
|
FAIL_EXIT1 ("create_temp_file: %m");
|
2016-09-17 04:44:50 +08:00
|
|
|
break;
|
|
|
|
}
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
files[nfiles] = fd;
|
2016-09-17 04:44:50 +08:00
|
|
|
}
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
TEST_VERIFY_EXIT (nfiles != 0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
posix_spawn_file_actions_t a;
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
TEST_COMPARE (posix_spawn_file_actions_init (&a), 0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
Fix hardcoded /tmp paths in testing (bug 13888).
As noted in bug 13888, and as I noted previously in
<https://sourceware.org/ml/libc-alpha/2000-10/msg00111.html>, various
tests used hardcoded paths in /tmp, so posing issues for simultaneous
test runs from different build directories.
This patch fixes such uses of hardcoded file names to put them in the
build directory instead (in the case of stdio-common/bug5 the file
names are changed as well, to avoid a conflict with the name bug5.out
also used for the automatic test output redirection). It also fixes
test-installation.pl likewise (that was using filenames with $$ in
them rather than strictly hardcoded names, but that's still not good
practice for temporary file naming).
Note that my list of files changed is not identical to that in bug
13888. I added tst-spawn3.c and test-installation.pl, and removed
some tests that seem to me (now) to create temporary files securely
(simply using /tmp is not itself a problem if the temporary files are
handled properly with mkstemp; I haven't checked whether those tests
used to do things insecurely). conformtest is not changed because the
makefiles always pass a --tmpdir option so the /tmp default is
irrelevant, and for the same reason there is no actual problem with
nptl/tst-umask1.c because again the makefiles always override the
default.
nptl/sockperf.c is ignored because there is no code to run it;
probably that file should actually be removed.
Some tests use the mktemp function, but I think they all use it in a
way that *is* secure (for generating names for directories / sockets /
fifos / symlinks, where the operation using the name will not follow
symlinks and so there is no potential for a symlink attack on the
account running the testsuite).
Some tests use the tmpnam function to generate temporary file names.
This is in principle insecure, but not addressed by this patch (I
consider it a separate issue from the fully hardcoded paths).
Tested for x86_64.
[BZ #13888]
* posix/Makefile (CFLAGS-tst-spawn3.c): New variable.
* posix/tst-spawn3.c (do_test): Put tst-spwan3.pid in OBJPFX, not
/tmp.
* scripts/test-installation.pl: Put temporary files in build
directory, not /tmp.
* stdio-common/Makefile (CFLAGS-bug3.c): New variable.
(CFLAGS-bug4.c): Likewise.
(CFLAGS-bug5.c): Likewise.
(CFLAGS-test-fseek.c): Likewise.
(CFLAGS-test-popen.c): Likewise.
(CFLAGS-test_rdwr.c): Likewise.
* stdio-common/bug3.c (main): Put temporary file in OBJPFX, not
/tmp.
* stdio-common/bug4.c (main): Likewise.
* stdio-common/bug5.c (main): Likewise.
* stdio-common/test-fseek.c (TESTFILE): Likewise.
* stdio-common/test-popen.c (do_test): Likewise.
* stdio-common/test_rdwr.c (main): Likewise.
2018-06-27 05:48:48 +08:00
|
|
|
/* Executes a /bin/sh echo $$ 2>&1 > ${objpfx}tst-spawn3.pid . */
|
|
|
|
const char pidfile[] = OBJPFX "tst-spawn3.pid";
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
TEST_COMPARE (posix_spawn_file_actions_addopen (&a, STDOUT_FILENO, pidfile,
|
|
|
|
O_WRONLY| O_CREAT | O_TRUNC,
|
|
|
|
0644),
|
|
|
|
0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
TEST_COMPARE (posix_spawn_file_actions_adddup2 (&a, STDOUT_FILENO,
|
|
|
|
STDERR_FILENO),
|
|
|
|
0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
/* Since execve (called by posix_spawn) might require to open files to
|
|
|
|
actually execute the shell script, setup to close the temporary file
|
|
|
|
descriptors. */
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
int maxnfiles =
|
|
|
|
#ifdef TST_SPAWN_PIDFD
|
|
|
|
/* The sparing file descriptor will be returned as the pid descriptor,
|
|
|
|
otherwise clone fail with EMFILE. */
|
|
|
|
nfiles - 1;
|
|
|
|
#else
|
|
|
|
nfiles;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
for (int i=0; i<maxnfiles; i++)
|
|
|
|
TEST_COMPARE (posix_spawn_file_actions_addclose (&a, files[i]), 0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
char *spawn_argv[] = { (char *) _PATH_BSHELL, (char *) "-c",
|
|
|
|
(char *) "echo $$", NULL };
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
PID_T_TYPE pid;
|
|
|
|
|
|
|
|
{
|
|
|
|
int r = POSIX_SPAWN (&pid, _PATH_BSHELL, &a, NULL, spawn_argv, NULL);
|
|
|
|
if (r == ENOSYS)
|
|
|
|
FAIL_UNSUPPORTED ("kernel does not support CLONE_PIDFD clone flag");
|
|
|
|
#ifdef TST_SPAWN_PIDFD
|
|
|
|
TEST_COMPARE (r, EMFILE);
|
|
|
|
|
|
|
|
/* Free up one file descriptor, so posix_spawn_pidfd_ex can return it. */
|
|
|
|
xclose (files[nfiles-1]);
|
|
|
|
nfiles--;
|
|
|
|
r = POSIX_SPAWN (&pid, _PATH_BSHELL, &a, NULL, spawn_argv, NULL);
|
|
|
|
#endif
|
|
|
|
TEST_COMPARE (r, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
siginfo_t sinfo;
|
|
|
|
TEST_COMPARE (WAITID (P_PID, pid, &sinfo, WEXITED), 0);
|
|
|
|
TEST_COMPARE (sinfo.si_code, CLD_EXITED);
|
|
|
|
TEST_COMPARE (sinfo.si_status, 0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
/* Close the temporary files descriptor so it can check posix_spawn
|
|
|
|
output. */
|
|
|
|
for (int i=0; i<nfiles; i++)
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
xclose (files[i]);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
int pidfd = xopen (pidfile, O_RDONLY, 0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
char buf[INT_BUFSIZE_BOUND (pid_t)];
|
|
|
|
ssize_t n = read (pidfd, buf, sizeof (buf));
|
|
|
|
TEST_VERIFY (n < sizeof buf && n >= 0);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
xunlink (pidfile);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
/* We only expect to read the PID. */
|
|
|
|
char *endp;
|
|
|
|
long int rpid = strtol (buf, &endp, 10);
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
TEST_VERIFY (*endp == '\n' && endp != buf);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
posix: Add pidfd_spawn and pidfd_spawnp (BZ 30349)
Returning a pidfd allows a process to keep a race-free handle for a
child process, otherwise, the caller will need to either use pidfd_open
(which still might be subject to TOCTOU) or keep the old racy interface
base on pid_t.
To correct use pifd_spawn, the kernel must support not only returning
the pidfd with clone/clone3 but also waitid (P_PIDFD) (added on Linux
5.4). If kernel does not support the waitid, pidfd return ENOSYS.
It avoids the need to racy workarounds, such as reading the procfs
fdinfo to get the pid to use along with other wait interfaces.
These interfaces are similar to the posix_spawn and posix_spawnp, with
the only difference being it returns a process file descriptor (int)
instead of a process ID (pid_t). Their prototypes are:
int pidfd_spawn (int *restrict pidfd,
const char *restrict file,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict],
char *const envp[restrict])
int pidfd_spawnp (int *restrict pidfd,
const char *restrict path,
const posix_spawn_file_actions_t *restrict facts,
const posix_spawnattr_t *restrict attrp,
char *const argv[restrict_arr],
char *const envp[restrict_arr]);
A new symbol is used instead of a posix_spawn extension to avoid
possible issues with language bindings that might track the return
argument lifetime. Although on Linux pid_t and int are interchangeable,
POSIX only states that pid_t should be a signed integer.
Both symbols reuse the posix_spawn posix_spawn_file_actions_t and
posix_spawnattr_t, to void rehash posix_spawn API or add a new one. It
also means that both interfaces support the same attribute and file
actions, and a new flag or file action on posix_spawn is also added
automatically for pidfd_spawn.
Also, using posix_spawn plumbing allows the reusing of most of the
current testing with some changes:
- waitid is used instead of waitpid since it is a more generic
interface.
- tst-posix_spawn-setsid.c is adapted to take into consideration that
the caller can check for session id directly. The test now spawns
itself and writes the session id as a file instead.
- tst-spawn3.c need to know where pidfd_spawn is used so it keeps an
extra file description unused.
Checked on x86_64-linux-gnu on Linux 4.15 (no CLONE_PIDFD or waitid
support), Linux 5.4 (full support), and Linux 6.2.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
2023-08-25 00:42:18 +08:00
|
|
|
TEST_COMPARE (rpid, sinfo.si_pid);
|
2016-09-17 04:44:50 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
2017-04-25 02:48:01 +08:00
|
|
|
|
|
|
|
#include <support/test-driver.c>
|