SymbX/systemd
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
84,907 Commits 2 Branches 1 Tag
2d6dd692beb67b9167d8f7eff91c83d8c9fcdc4e
Commit Graph
84907 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Yu Watanabe 2d6dd692be cgroup-util: drop cgroup v1 support in cg_get_path() 
We have dropped cgroup v1 support in v258. Let's assume we are running
on cgroup v2 and drop unused legacy code.
 2025-11-17 21:30:29 +09:00
Yu Watanabe f5a48af34b cgroup-util: use cg_is_available() in cg_kill_supported() 2025-11-17 21:30:29 +09:00
Yu Watanabe 57c16d344b cgroup-util: drop cgroup v1 support from cg_mask_supported_subtree() 
We have dropped cgroup v1 support in v258. Let's drop unused legacy code.
 2025-11-17 21:30:29 +09:00
Yu Watanabe 0ab90015e0 tree-wide: replace cg_get_path_and_check() with cg_get_path() 
We have dropped cgroup v1 support in v258. When running on cgroup v2,
cg_get_path_and_check() with SYSTEMD_CGROUP_CONTROLLER as controller is
equivalent with checking if we are running on cgroup v2 and then
cg_get_path(). As we can assume we are running on cgroup v2, then the
check is not necessary anymore, thus we can replace
cg_get_path_and_check() with cg_get_path().
 2025-11-17 21:30:29 +09:00
Yu Watanabe b525a72f7b cgroup-util: drop cgroup v1 support from cg_pid_get_path() 
We have dropped cgroup v1 support in v258. Let's drop legacy code.
Then, we can drop 'controller' argument from cg_pid_get_path() and
cg_pidref_get_path().
 2025-11-17 21:30:29 +09:00
Yu Watanabe 6475926a59 cgroup-util: drop 'controller' argument from cg_path_open() 
Nowadays it always takes SYSTEMD_CGROUP_CONTROLLER as controller.
Let's drop it.

No functional change, just refactoring.
 2025-11-17 21:30:29 +09:00
Yu Watanabe 4dbf06bd85 cgroup-util: drop 'controller' argument from cg_set_attribute(), cg_get_attribute() and friends 
Non-null controller arguments are always ignored when running on cgroup v2.
Let's drop the argument.

No functional change, just refactoring.
 2025-11-17 21:30:29 +09:00
Yu Watanabe 4d1badbbc4 cgroup-util: drop 'controller' argument from cg_enumerate_subgroups() 
Nowadays it always takes SYSTEMD_CGROUP_CONTROLLER as controller.
Let's drop it.

No functional change, just refactoring.
 2025-11-17 21:30:29 +09:00
Yu Watanabe f8de2107a7 cgroup-util: drop 'controller' argument from cg_enumerate_processes() 
Nowadays it always takes SYSTEMD_CGROUP_CONTROLLER as controller.
Let's drop it.

No functional change, just refactoring.
 2025-11-17 21:30:29 +09:00
Yu Watanabe 5169b1c5a7 cgroup-show: drop 'controller' argument from show_cgroup() and show_cgroup_and_extra() 
Nowadays these always take SYSTEMD_CGROUP_CONTROLLER as controller.
Let's drop it.

No functional change, just refactoring.
 2025-11-17 21:30:29 +09:00
Yu Watanabe 4365351b93 cgroup-util: introduce cg_is_available() and check it in tests 
Now most of our code does not support cgroup v1. Let's skip test cases
if we are running on cgroup v1.
 2025-11-17 21:29:51 +09:00
Luca Boccassi bc154d4c3e test: skip TEST-64-UDEV-STORAGE-simultaneous_events on Debian 
It consistently fails with the new 6.17 kernel and nobody can
figure out why, so just disable it for now as it's just noise

Fixes https://github.com/systemd/systemd/issues/39552
 2025-11-17 12:17:24 +00:00
Lennart Poettering 43939a5bbe NEWS: first prep for v259 2025-11-17 12:57:37 +01:00
Luca Boccassi b186ce49de Chores for RC1 (#39757) 2025-11-17 10:53:15 +00:00
Zbigniew Jędrzejewski-Szmek 651fdde1cd Add experimental musl support (#38825) 
This adds experimental support of building systemd with musl. This
allows to build systemd with musl and run most of the unit tests.
Running integration tests is not yet supported, however. Hopefully, this
should be a good starting point to support musl.

This requires musl-1.2.5 with
https://github.com/kraj/musl/commit/fde29c04adbab9d5b081bf6717b5458188647f1c.
The patch is already backported to Alpine/postmarketOS's musl package
since musl-1.2.5-r11. See
https://gitlab.alpinelinux.org/alpine/aports/-/commit/333e50c2052c74bb372f821fdeea2e6814f2fe7a.
 2025-11-17 11:14:47 +01:00
Luca Boccassi b4fb69d1d7 man: regenerate man/rules/meson.build 
ninja -C build update-man-rules
 2025-11-17 08:59:53 +00:00
Luca Boccassi 79ce4f577f Translations: update systemd-po 
ninja -C build systemd-update-po
 2025-11-17 08:58:51 +00:00
Luca Boccassi 6006b55e8a Translations: update systemd-pot 
ninja -C build systemd-pot
 2025-11-17 08:57:57 +00:00
Luca Boccassi 5b81a76313 Update syscalls table 
ninja -C build update-syscall-tables update-syscall-header
 2025-11-17 08:56:39 +00:00
Luca Boccassi 8ae29e3fea NEWS: update contributors list 2025-11-17 08:55:28 +00:00
Luca Boccassi 0c34bec7bb Update hwdb 
ninja -C build update-hwdb
 2025-11-17 08:54:08 +00:00
Lennart Poettering e9ce56c6e6 update TODO 2025-11-17 08:54:06 +01:00
Armin Wolf 2fc4278b96 keymap: Ignore brightness keys on Dell Inspiron 3505 to avoid double events 
On the Dell Inspiron 3505 both the atkbd and acpi-video input devices report
an event for pressing the brightness up / down keys, resulting in user
space seeing double events and increasing / decreasing the brightness 2 steps
for each keypress.

Fix this by adding the device to the already existing list of Dell
devices that suffer from the same problem.

Signed-off-by: Armin Wolf <W_Armin@gmx.de>
 2025-11-17 13:37:26 +09:00
Yu Watanabe f115763f19 musl: ci: add build test and unit tests 2025-11-17 12:19:22 +09:00
Yu Watanabe 07e437f569 musl: glob-util: filter out . and .. even if GLOB_ALTDIRFUNC is not supported 
musl neither support GLOB_ALTDIRFUNC nor GLOB_BRACE.
Let's make safe_glob() work even when GLOB_ALTDIRFUNC is not supported.
Currently, GLOB_BRACE is simply ignored when it is not supported.
 2025-11-17 12:19:22 +09:00
Emil Renner BerthingandYu Watanabe 3bc03c67f0 musl: add fallback parse_printf_format() implementation 
musl does not provide parse_printf_format(). Let's introduce a fallback
method.

Co-authored-by: Yu Watanabe <watanabe.yu+github@gmail.com>
 2025-11-17 12:19:22 +09:00
Yu Watanabe bf9bc5beb0 musl: meson: add libutmps support 
musl only provides fake functions, hence even with -Dutmp=true,
utmp/wtmp functionalities do not work. Let's allow to build with
libutmps.
 2025-11-17 12:18:44 +09:00
Yu Watanabe 3ae7d8fd87 musl: utmpx: add several missing definitions 2025-11-17 11:27:07 +09:00
Lennart Poettering 1240ba1807 iovec-util: drop redundant () 2025-11-16 23:36:11 +01:00
Lennart Poettering 08d75cfb99 dissect-image,execute: switch root hash/root hash sig storage to struct iovec 
let's go one step further on our iovec'ification journey.

No change in behaviour, just rework to make struct iovec used
everywhere.
 2025-11-16 23:36:11 +01:00
Yu Watanabe ddc9677c4f test-network: several tweaks (#39752) 2025-11-17 06:39:45 +09:00
Yu Watanabe d7cf08b735 test-network: drop two overly verbose outputs 
Follow-up for 7f9c0c31d2.
 2025-11-16 22:15:44 +09:00
Yu Watanabe 3632d7322f test-network: set TimeoutStopFailureMode=abort when running with sanitizers 
Hopefully this provides useful data for issue #39631.
 2025-11-16 22:15:43 +09:00
Yu Watanabe c43585bc30 Revert "test-network: disable several more sandbox features when running with sanitizers" 
This reverts commit 454ce42305.

Unfortunately, it does not work.
 2025-11-16 22:15:43 +09:00
David Tardon 8fc02a8cfd logind-dbus: avoid assertion on HandlePowerKey=factory-reset 
The factory-reset action is handled together with the various
sleep/shutdown actions, but its not either one.

Fixes #39385
 2025-11-16 17:00:53 +09:00
Chris Down 77a92e97eb tree-wide: several cleanups (#39746) 2025-11-16 15:40:39 +08:00
Chris Down 92cbc95e7a network/dhcp: several follow-ups for recent change (#39744) 
Follow-ups for 7f9c0c31d2
 2025-11-16 15:39:56 +08:00
Yu Watanabe 4f9f6a8517 resolve: voidify one function call 
Fixes CID#1643235.
 2025-11-16 14:06:26 +09:00
Yu Watanabe a112c7646b tree-wide: swap arguments for streq() 
We usually set a variable first, then constant.
 2025-11-16 14:01:56 +09:00
Yu Watanabe 33b2dc77bd test-network: migrate to use ASSERT_OK() and friends 2025-11-16 14:01:39 +09:00
Yu Watanabe 5a943bae12 network-internal: do not save partially parsed values 2025-11-16 14:01:28 +09:00
Yu Watanabe 8c236eca78 resolve: fix typo 
Follow-up for 8209f4adcd.
 2025-11-16 11:13:56 +09:00
Yu Watanabe e7254f9f4e efivars: fix typo 
Follow-up for ab69a04600.
 2025-11-16 11:12:10 +09:00
Yu Watanabe af70b7521f sd-dhcp-lease: use the last hostname option when multiple hostname is specified 2025-11-16 10:50:34 +09:00
Yu Watanabe a6c61b4adc sd-dhcp-lease: drop ret_ prefixes from the parsers 2025-11-16 10:49:59 +09:00
Yu Watanabe f28bc48434 network: unify log messages 2025-11-16 10:44:37 +09:00
Daan De Meyer 6de1c68924 sd-event: several follow-ups for recent change (#39743) 2025-11-15 22:47:42 +01:00
Yu Watanabe de8ea7e219 test: add missing assertion 
Follow-up for 5a5cb6ba50.
Fixes CID#1643217.
 2025-11-16 05:51:22 +09:00
Yu Watanabe 8e4ef4a18c sd-event: do not ignore result of callbacks 
Follow-up for 4c8b6d636c and
6aff6d3ffc.
Fixes CID#1643218 and friends.
 2025-11-16 05:44:35 +09:00
Yu Watanabe c9fc7eb87e nspawn: Prevent invalid UIDs propagating in bind mounts (#39729) 
Commit 88fce09026 modified the
mount_bind() function, causing it to perform arithmetic on the uid_shift
parameter. However, it performs this arithmetic even when uid_shift was
UID_INVALID, which was not intended. This typically occurred when
mount_custom() was called for a simple bind mount without user
namespaces (and thus no rootidmap mount option).

This arithmetic (e.g., uid_shift + m->destination_uid) then wraps
around, resulting in the invalid ID 4294967295 ((uid_t)-1).

This bug manifests for users running systemd-nspawn with
--link-journal=host and --volatile=yes (but without --private-users),
causing systemd-tmpfiles to fail.

Make mount_bind() robust by checking if uid_shift is valid before using
it in arithmetic. If it is UID_INVALID, it defaults to a shift of 0 for
the ownership calculation, restoring correct behavior for plain bind
mounts while preserving the intended logic for ID-mapped mounts.

Fixes: #39714
 2025-11-16 04:46:18 +09:00