Luca Boccassi [Mon, 11 Aug 2025 14:33:35 +0000 (15:33 +0100)]
sd-stub: use memory proto if available and set kernel memory to RX with NX_COMPAT
When NX_COMPAT gets enabled, firmwares will enforce that executable
memory is either writable or executable.
This needs kernel compatibility, when it will happen the kernel will
have the NX_COMPAT bit set. If it is, set the memory buffer to RO.
Note that this must be undone on failure, as EDK2 in some configurations
overwrites memory ranges that are returned with FreePages() with a
fixed pattern, so if the pages are RO it will crash.
This is only an issue with the new custom PE loader, as LoadImage()
and StartImage() will always do the right thing automatically.
Luca Boccassi [Thu, 14 Aug 2025 16:22:30 +0000 (17:22 +0100)]
chase: invert CHASE_NO_AUTOFS and only set it where needed
Since https://github.com/systemd/systemd/commit/c5de7b14ae2e08d267d8d75bc88934ac6aa7dcd6
file searching implies a new mount api syscall by default,
to trigger automounts.
This is problematic in NSS plugins, as they are dlopen'ed inside
processes by glibc, for two reasons.
First of all, potentially searching on a networked filesystem
automount could lead to nasty surprises, such as the process
responsible for setting up the network filesystem trying to
search on that same filesystem.
More importantly, the new mount api syscall was never part of
the filesystem seccomp filter that we provide by default, and
given mounting/remounting/bind mounting is one of the possible
ways to bypass sandboxing it is very likely not allowed when
custom filters are used in sandboxed processes, if they don't
need to do these operations otherwise.
The filesystem seccomp mask we provide has been updated, however
this only takes effect on the next restart of a service. When
systemd is upgraded via a package upgrade, the new nss plugin is
installed and will be immediately dlopen'ed by glibc when needed,
without waiting for the process to restart, which means the existing
seccomp filter applies, causing the filter to trigger.
Given it's not really possible for any arbitrary program to
predict which NSS modules glibc will load, given programs do not
configure that and instead nsswitch is set up by the sysadmin,
it's impossible to handle at each process level. It's also not
possible to know when it will be triggered, given the plugin
is not linked in each binary tools like need-restart cannot
even pre-emptively restart services that may be affected.
This means in practice, upgrading from systemd << v258 to >= v258
requires a reboot to avoid either subtle or catastrophic system
failures.
By avoiding to trigger automounts in nss-systemd we can avoid
both issues.
Note that this happens only when the userdbd service is not running,
as otherwise nss-systemd will go through the varlink IPC, rather than
doing the searches in-process.
So invert CHASE_NO_AUTOFS to CHASE_AUTOFS and set it in the places where
we do want to trigger automounts, like looking for the ESP.
Luca Boccassi [Tue, 19 Aug 2025 11:32:00 +0000 (12:32 +0100)]
test: also edit /etc/os-release if it's not a symlink when patching /usr/lib/os-release (#38628)
mkosi patches up /etc/os-release to add local IDs and fixup certain
issues, so when tests patch /usr/lib/ on the fly, copy to the version in
/etc/ too to avoid test failures when querying
6370s 10/98 systemd:integration-tests / TEST-07-PID1 FAIL 31.03s exit
status 1
6370s 25/98 systemd:integration-tests / TEST-29-PORTABLE FAIL 12.76s
exit status 1
6370s 33/98 systemd:integration-tests / TEST-43-PRIVATEUSER-UNPRIV FAIL
6.57s exit status 1
6370s 37/98 systemd:integration-tests / TEST-50-DISSECT FAIL 16.97s exit
status 1
This is particularly an issue when running these tests on debian unstable,
where mkosi has to fixup os-release to make it valid and avoid further
breakages:
importd: accept a single space as SHA256SUMS separator
The SHA256SUMS files provided by https://images.linuxcontainers.org/
are slightly non-conforming, insted of using " *" or " " as separator
between hash and file name they use " ". Let's accept that too, in the
interest of maximizing compatibility.
Currently translated at 100.0% (264 of 264 strings)
Co-authored-by: Fco. Javier F. Serrador <fserrador@gmail.com>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/main/es/
Translation: systemd/main
Li Tian [Mon, 18 Aug 2025 21:43:41 +0000 (05:43 +0800)]
ukify: rstrip and escape binary null characters from 'inspect' output (#38607)
SBAT section of UKI may contain \u000 null characters. Rstrip them, and if there's anything left in the middle,
escape them so they are displayed as text.
tree-wide: don't play games with alignment around file_handle
The payload of a file_handle structure is not 64bit aligned. So far used
_alignas_() to align it to 64bit as a whole, which by accident has the
side-effect that the payload ends up being aligned to 64bit too, but
this is ugly, because it's really just an accident...
Let's do this properly, and just use proper unaligned 64bit reads to
access the field, and do not assume aligning the structure as a whole
also aligns the payload part of it.
Yu Watanabe [Mon, 18 Aug 2025 06:46:04 +0000 (15:46 +0900)]
ptyfwd: do not try to read from PTYForward.input_fd when read-only mode
Fixes the following error message (the last line):
```
[FAILED] Failed to start TEST-60-MOUNT-RATELIMIT.service.
Sending SIGTERM to remaining processes...
Sending SIGKILL to remaining processes...
All filesystems, swaps, loop devices, MD devices and DM devices detached.
Exiting container.
Failed to read from pty input fd: Bad file descriptor
```
Luca Boccassi [Sun, 17 Aug 2025 23:34:22 +0000 (00:34 +0100)]
vmspawn: fix --smbios
The file used to pass credentials gets created and then immediately deleted:
$ systemd-vmspawn -i image.raw -s "io.systemd.credential.binary:tty.serial.hvc0.agetty.autologin=cm9vdA=="
░ Spawning VM opensuse-2025081621.1 on /tmp/image.raw.
░ Press Ctrl-] three times within 1s to kill VM.
Not overwriting existing state file.
Listening on /run/user/1000/systemd/vmspawn.1c00857c6a3dc2c7/tpm.sock as 3.
qemu-system-x86_64: -smbios type=11,path=/var/tmp/vmspawn-smbios-Hizb4A/.#smbios11e5a842e77d7b4b68: Could not open '/var/tmp/vmspawn-smbios-Hizb4A/.#smbios11e5a842e77d7b4b68': No such file or directory
Yu Watanabe [Sun, 17 Aug 2025 16:15:33 +0000 (01:15 +0900)]
README: drop one FIXME comment
Most compat glue has been already removed, except for several cgroup v1
specific codes. It is too late to remove the remaining things before v258.
Let's remove them after v258.
Luca Boccassi [Sat, 16 Aug 2025 22:56:11 +0000 (23:56 +0100)]
core: fix crash on audit callback
When check_access() was added, the callback data parameter
was changed from a pointer to a double pointer, resulting
in a crash when it is accessed when logging an error:
#0 __internal_syscall_cancel (a1=a1@entry=0, a2=a2@entry=0,
a3=a3@entry=140726176497168, a4=a4@entry=4, a5=a5@entry=0,
a6=a6@entry=0, nr=247) at ./nptl/cancellation.c:44
#1 0x00007f5d0ec996ad in __syscall_cancel (a1=a1@entry=0,
a2=a2@entry=0, a3=a3@entry=140726176497168, a4=a4@entry=4,
a5=a5@entry=0, a6=a6@entry=0, nr=247) at ./nptl/cancellation.c:75
#2 0x00007f5d0ed047ab in __waitid (idtype=idtype@entry=P_ALL,
id=id@entry=0, infop=infop@entry=0x7ffd5dc2be10,
options=options@entry=4) at ../sysdeps/unix/sysv/linux/waitid.c:29
#3 0x00007f5d0f044412 in freeze () at ../src/basic/process-util.c:2039
#4 0x00005568f181bc2a in freeze_or_exit_or_reboot () at
../src/core/crash-handler.c:55
#5 0x00005568f181be82 in crash (sig=<optimized out>, siginfo=<optimized
out>, context=<optimized out>) at ../src/core/crash-handler.c:184
#6 <signal handler called>
#7 __strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:76
#8 0x00007f5d0ec6e300 in __printf_buffer (buf=buf@entry=0x7ffd5dc2ca90,
format=0x7f5d0f196e60 "%s: Failed to acquire credentials: %m",
ap=0x7ffd5dc2d3d0, mode_flags=2) at
./stdio-common/vfprintf-process-arg.c:435
#9 0x00007f5d0ec91daf in __vsnprintf_internal
(string=string@entry=0x7ffd5dc2cb70 "", maxlen=maxlen@entry=2048,
format=format@entry=0x7f5d0f196e60 "%s: Failed to acquire credentials:
%m", args=args@entry=0x7ffd5dc2d3d0, mode_flags=mode_flags@entry=2)
at ./libio/vsnprintf.c:96
#10 0x00007f5d0ed27044 in ___vsnprintf_chk (s=s@entry=0x7ffd5dc2cb70 "",
maxlen=maxlen@entry=2048, flag=flag@entry=1, slen=slen@entry=2048,
format=format@entry=0x7f5d0f196e60 "%s: Failed to acquire credentials:
%m", ap=ap@entry=0x7ffd5dc2d3d0)
at ./debug/vsnprintf_chk.c:34
#11 0x00007f5d0f02de59 in vsnprintf (__s=0x7ffd5dc2cb70 "", __n=2048,
__fmt=0x7f5d0f196e60 "%s: Failed to acquire credentials: %m",
__ap=0x7ffd5dc2d3d0) at /usr/include/x86_64-linux-gnu/bits/stdio2.h:100
#12 log_internalv (level=7, error=-9, file=0x7f5d0f196643
"src/libsystemd/sd-varlink/sd-varlink.c", line=2853, func=0x7f5d0f1d5ca0
<__func__.62> "sd_varlink_get_peer_uid", format=0x7f5d0f196e60 "%s:
Failed to acquire credentials: %m",
ap=0x7ffd5dc2d3d0) at ../src/basic/log.c:865
#13 0x00007f5d0f02ded5 in log_internalv (level=<optimized out>,
error=<optimized out>, file=<optimized out>, line=<optimized out>,
func=<optimized out>, format=<optimized out>, ap=0x7ffd5dc2d3d0) at
../src/basic/log.c:868
#14 0x00007f5d0f02df67 in log_internal (level=<optimized out>,
error=<optimized out>, file=<optimized out>, line=<optimized out>,
func=<optimized out>, format=<optimized out>) at ../src/basic/log.c:882
#15 0x00007f5d0f10a135 in sd_varlink_get_peer_uid (v=0x7f5d0f5ab110
<__func__.44>, ret=ret@entry=0x7ffd5dc2d4f0) at
../src/libsystemd/sd-varlink/sd-varlink.c:2853
#16 0x00007f5d0f50c29e in audit_callback (auditdata=0x7ffd5dc2d698,
cls=<optimized out>, msgbuf=0x55692366e77d "", msgbufsize=995) at
../src/core/selinux-access.c:65
#17 0x00007f5d0f716079 in avc_suppl_audit (ptr=0x7ffd5dc2d698, class=95,
buf=<optimized out>, len=<optimized out>) at ./src/avc_internal.h:101
#18 avc_audit (ssid=0x5569237f2890, tsid=0x556922d7f4b0,
tclass=<optimized out>, requested=<optimized out>, avd=<optimized out>,
result=0, a=0x7ffd5dc2d698) at ./src/avc.c:721
#19 0x00007f5d0f716367 in avc_has_perm (ssid=0x5569237f2890,
tsid=0x556922d7f4b0, tclass=tclass@entry=95, requested=4,
aeref=aeref@entry=0x0, auditdata=auditdata@entry=0x7ffd5dc2d698) at
./src/avc.c:836
#20 0x00007f5d0f718b0a in selinux_check_access
(scon=scon@entry=0x55692384cbc0 "system_u:system_r:policykit_t:s0",
tcon=tcon@entry=0x556922c98a20
"system_u:object_r:systemd_networkd_unit_t:s0",
class=class@entry=0x7f5d0f580b9e "service",
perm=perm@entry=0x7f5d0f580cc0 "status",
aux=aux@entry=0x7ffd5dc2d698) at ./src/checkAccess.c:64
#21 0x00007f5d0f50bf7e in check_access (scon=0x55692384cbc0
"system_u:system_r:policykit_t:s0", tcon=0x556922c98a20
"system_u:object_r:systemd_networkd_unit_t:s0", tclass=0x7f5d0f580b9e
"service", permission=permission@entry=0x7f5d0f580cc0 "status",
audit_info=<optimized out>, audit_info@entry=0x7ffd5dc2d720,
error=error@entry=0x7ffd5dc2d880) at ../src/core/selinux-access.c:229
#22 0x00007f5d0f5100a1 in mac_selinux_access_check_bus_internal
(message=<optimized out>, unit=<optimized out>,
permission=0x7f5d0f580cc0 "status", function=0x7f5d0f5ab110
<__func__.44> "method_get_unit_by_pidfd", error=0x7ffd5dc2d880)
at ../src/core/selinux-access.c:329
#23 0x00007f5d0f4a024b in method_get_unit_by_pidfd
(message=0x5569236d9010, userdata=<optimized out>, error=0x7ffd5dc2d880)
at ../src/core/dbus-manager.c:657
#24 0x00007f5d0f0c9bd0 in method_callbacks_run (bus=0x5569238684e0,
m=0x5569236d9010, c=<optimized out>, require_fallback=false,
found_object=0x7ffd5dc2d947) at ../src/libsystemd/sd-bus/bus-objects.c:413
#25 object_find_and_run (bus=bus@entry=0x5569238684e0,
m=m@entry=0x5569236d9010, p=<optimized out>,
require_fallback=require_fallback@entry=false,
found_object=found_object@entry=0x7ffd5dc2d947) at
../src/libsystemd/sd-bus/bus-objects.c:1323
#26 0x00007f5d0f0cafa2 in bus_process_object (bus=0x5569238684e0,
m=0x5569236d9010) at ../src/libsystemd/sd-bus/bus-objects.c:1443
#27 0x00007f5d0f0d8c3e in process_message (bus=0x5569238684e0,
m=0x5569236d9010) at ../src/libsystemd/sd-bus/sd-bus.c:3006
#28 process_running (bus=0x5569238684e0, ret=0x0) at
../src/libsystemd/sd-bus/sd-bus.c:3048
#29 bus_process_internal (bus=bus@entry=0x5569238684e0,
ret=ret@entry=0x0) at ../src/libsystemd/sd-bus/sd-bus.c:3275
#30 0x00007f5d0f0d9099 in sd_bus_process (bus=bus@entry=0x5569238684e0,
ret=ret@entry=0x0) at ../src/libsystemd/sd-bus/sd-bus.c:3302
#31 0x00007f5d0f0db3ec in io_callback (s=<optimized out>, fd=<optimized
out>, revents=<optimized out>, userdata=0x5569238684e0) at
../src/libsystemd/sd-bus/sd-bus.c:3643
#32 0x00007f5d0f0a53d9 in source_dispatch (s=s@entry=0x5569236dea60) at
../src/libsystemd/sd-event/sd-event.c:4163
#33 0x00007f5d0f0a563d in sd_event_dispatch (e=<optimized out>,
e@entry=0x5569232f6c00) at ../src/libsystemd/sd-event/sd-event.c:4782
#34 0x00007f5d0f0a6d38 in sd_event_run (e=<optimized out>,
timeout=18446744073709551615) at ../src/libsystemd/sd-event/sd-event.c:4843
#35 0x00007f5d0f4f7871 in manager_loop (m=m@entry=0x5569232f8250) at
../src/core/manager.c:3310
#36 0x00005568f181517d in invoke_main_loop (m=0x5569232f8250,
saved_rlimit_nofile=0x7ffd5dc2dcb0, saved_rlimit_memlock=0x7ffd5dc2dca0,
ret_retval=<synthetic pointer>, ret_fds=0x7ffd5dc2dc78,
ret_switch_root_dir=<synthetic pointer>,
ret_switch_root_init=<synthetic pointer>,
ret_error_message=0x7ffd5dc2dc90) at ../src/core/main.c:2140
#37 main (argc=<optimized out>, argv=0x7ffd5dc2dfe8) at
../src/core/main.c:3351
Yu Watanabe [Sat, 16 Aug 2025 16:00:37 +0000 (01:00 +0900)]
core/service: do not reset watchdong when unit is frozen
Even watchdog for a service is stopped when freezing the unit is
requested, sd-notify message WATCHDOG=1 or friends may come after
that due to ordering of event priority. In that case,
service_reset_watchdog() is called for frozen unit and thus previously
watchdog was reenabled.
Yu Watanabe [Fri, 15 Aug 2025 04:57:51 +0000 (13:57 +0900)]
signal-util: do not abort when sigprocmask() failed
BLOCK_SIGNALS() is also used in nss modules. If an application is
running with a too strict seccomp loads our nss modules, then the
assertion may be triggered.
Yu Watanabe [Fri, 15 Aug 2025 04:21:59 +0000 (13:21 +0900)]
hostname-setup: do not trigger assertion when uname() is prohibited by seccomp
gethostname_full() is used in nss-myhostname, and hence random
application may indirectly call it. When an application with a too strict
seccomp filter loads the nss module, the application may trigger the
assertion.
* 5598b7f579 fedora: be more persistent when guessing what rawhide could be
* cdd2d1570e Use apt-ftparchive instead of reprepro
* eeb4ce6302 fix dead/404 link
* 30a487d183 mkosi-tools: Drop systemd-boot-efi package
* ad4b4d2cbe Add debug logging for version reported by systemd tools
* 95f5c77fb7 mkosi-tools: move systemd-boot package to conf file matching older releases
* 7da22f33e0 README: clarify that companion tools can also be enabled from the git repo
* ec3fe91532 Drop microsecond resolution for datetime.now()
* 9f7a53b687 mkosi-initrd: install raid rule with 70 prefix
* 32c3ff4677 ci: give a hint about possible fixes for failing reuse lints
* 489c5e9ecc build(deps): bump github/codeql-action from 3.29.2 to 3.29.5
Mate Kukri [Thu, 7 Aug 2025 16:28:58 +0000 (17:28 +0100)]
Reuse the parent_image handle and parent_loaded_image
- Reuse parent_image instead of allocating new ones. Firmware might cast
EFI_LOADED_IMAGE_PROTOCOL * to a larger struct causing issues
- Remove loaded image protocol installation and uninstallation which are no
longer required
Luca Boccassi [Tue, 12 Aug 2025 22:09:06 +0000 (23:09 +0100)]
ukify: drop NX bit from UKI if kernel doesn't have it
If the kernel is not NX_COMPAT ready (W^X memory compatible) then the
UKI should not be marked as NX_COMPAT ready either, as the kernel
section is the loadable code in the image.
While the sd-stub EFI code itself is NX ready, it is more useful
to think of it as one unit of execution together with the kernel
it embeds, as that's what it is used for.
keentux [Wed, 13 Aug 2025 07:52:12 +0000 (09:52 +0200)]
detect-virt: bare-metal GCE only for x86 and i386
From the previous changes, bare-metal support has been added by using
the `detect_vm_cpuid()` which works only for x86_64 and i386 architecture.
Do not use this change for other architectures to avoid wrong result of
the detect-virt tool.
Yu Watanabe [Wed, 13 Aug 2025 01:38:53 +0000 (10:38 +0900)]
core/cgroup: allow to set cgroup path for frozen unit
Otherwise, after 'systemctl daemon-reload' or 'daemon-reexec', frozen
units cannot gain cgroup paths and we cannot operate anything on them,
especially, we cannot thaw or stop them.
```
Aug 12 16:26:09 systemd[1]: wd.service: Job 1278 wd.service/stop finished, result=frozen
Aug 12 16:26:09 systemd[1]: Cannot stop frozen unit wd.service.
Aug 12 16:26:09 systemd[1]: wd.service: Cannot realize cgroup for frozen unit.
Aug 12 16:26:09 systemd[1]: Failed to realize cgroups for queued unit wd.service, ignoring: Device or resource busy
```
Luca Boccassi [Tue, 12 Aug 2025 22:59:15 +0000 (23:59 +0100)]
test-cgroup: cleanup test cgroup
One test cgroup gets left behind by the test, as it moves itself
into it. Move itself and back to the original cgroup at the end
and clean up.
This fixes a failure when running the test first as root, and then
as unprivileged (initial cleanup fails as the leftover test cgroup
is owned by root).
Matteo Croce [Tue, 12 Aug 2025 16:53:59 +0000 (18:53 +0200)]
core: suppress warning
Avoid definition of `exec_context_get_tty_for_pam` if pam support is
disabled, to avoid the following warning:
```
../src/core/exec-invoke.c:1231:12: warning: ‘exec_context_get_tty_for_pam’ defined but not used [-Wunused-function]
1231 | static int exec_context_get_tty_for_pam(const ExecContext *context, char **ret) {
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
```
Marcos Alano [Sat, 9 Aug 2025 10:52:27 +0000 (07:52 -0300)]
Enable KEY_PERFORMANCE key present on Linux 6.17
Note, this change does not require the kernel running on the host is
equal or newer than 6.17. But systemd-udevd needs to be built with the
kernel headers with KEY_PERFORMANCE, and the relevant kernel header is
already updated by the previous commit.
Found using linkchecker.
For virtiofsd, the man page is maintained upstream, but doesn't seem to be
available in any of the usual places. So let's link to the Debian version.
systemd.filter I have no idea what it is.
Vasiliy Kovalev [Sun, 10 Aug 2025 07:24:19 +0000 (10:24 +0300)]
hwdb: Add launch emoji keyboard mapping for Asus M1607KA
By default, pressing Fn+F8 maps the scancode to KEY_BLUETOOTH (in evtest,
MSC_SCAN 7e -> KEY_BLUETOOTH). Windows/the manufacturer may intercept the
same scancode to execute "Launch Emoji keyboard."
On Linux, we get the "raw" KEY_BLUETOOTH code, which is unacceptable.
prog1 is already reserved by default for launching MyAsus (a Windows
application) with the Fn+F12 combination, so we will use prog2.
Yu Watanabe [Fri, 8 Aug 2025 01:06:22 +0000 (10:06 +0900)]
test-network: also save the current state of socket units for resolved and stop them
Silence the following waring:
```
Stopping 'systemd-resolved.service', but its triggering units are still active:
systemd-resolved-varlink.socket, systemd-resolved-monitor.socket
```
logging: Improve logging messages related to NFTSet.
The 'NFTSet' directive in various units adds and removes entries in nftables
sets, it does not add or remove entire sets. The logging messages should
indicate that an entry was added or removed, not that a set was added or
removed.
Luca Boccassi [Wed, 6 Aug 2025 13:33:10 +0000 (14:33 +0100)]
test: use Europe/Helsinki instead of Europe/Kyiv in test-calendarspec
Europe/Kyiv was added somewhat recently. Use Europe/Helsinki which is
much older and thus works with older tzdata like version 2022a.
line 193: "2016-03-27 03:17:00" new_tz=:Europe/Kyiv
At: Sun 2016-03-27 03:17:00.000000 Europe
Assertion 'r == -ENOENT' failed at src/test/test-calendarspec.c:70, function _test_next(). Aborting.