There's a lot of code movement, but the actual changes are
straightforward. Previously, the program was not marked as public, so
the --help/--version interface wasn't tested.
Split the single directory fd that chaseat() used to take into two
separate
fds: a root_fd that sets the chroot boundary (symlinks may not escape
it,
absolute symlinks resolve relative to it), and a dir_fd that path
resolution
starts from. This makes the chroot semantics of chaseat() explicit at
every
call site instead of encoding them in the CHASE_AT_RESOLVE_IN_ROOT flag,
which is removed. It also decouples the starting directory from the root
boundary, so callers can descend from any inode inside the tree without
having to reopen the root separately.
XAT_FDROOT passed as root_fd means "no containment" (host root); as
dir_fd
it means "start at root_fd". For a smoother transition, AT_FDCWD is also
accepted as root_fd and treated as XAT_FDROOT. When root_fd points to a
directory that is actually the host root, it is normalized to XAT_FDROOT
up front so the existing shortcut path can kick in.
Absolute paths returned by chaseat() are now relative to root_fd, and
relative paths are relative to dir_fd. The result is absolute only when
there is no chroot boundary (root_fd is XAT_FDROOT), or when an absolute
symlink made resolution jump out of the dir_fd subtree; otherwise
callers
get a relative path they can feed straight back into an openat()-style
call against dir_fd. Specifically, when dir_fd == root_fd and we're not
operating on the host's root directory, we return a relative path even
if
we received an absolute path or resolved an absolute symlink to allow
passing the path directly to openat() style functions. We do this to not
have to go modify every caller of chaseat() to make sure they deal
properly
with any absolute paths they might receive. Only when root_fd != dir_fd
do
we have to return an absolute path to indicate that the path is relative
to
root_fd and not dir_fd.
The shortcut that skips the per-component walk is reworked around a new
chase_xopenat() helper that funnels CHASE_NOFOLLOW, CHASE_MUST_BE_* and
CHASE_TRIGGER_AUTOFS through xopenat_full()'s O_NOFOLLOW, O_DIRECTORY,
XO_REGULAR, XO_SOCKET and XO_TRIGGER_AUTOMOUNT flags. As a result these
flags no longer force us off the shortcut and can be dropped from
CHASE_NO_SHORTCUT_MASK, and the old openat_opath_with_automount() helper
goes away. A CHASE_MUST_BE_ANY alias is introduced for shortcut callers
(stat/access paths) that don't go through xopenat_full() and still need
to bail on those flags locally.
All *_and_* helpers built on top of chaseat() (chase_and_openat,
chase_and_opendirat, chase_and_statat, chase_and_accessat,
chase_and_fopenat_unlocked, chase_and_unlinkat,
chase_and_open_parent_at)
gain the same root_fd parameter, and every call site in the tree is
ported to the new signature.
The VERB definitions are done in order to retain the logical
presentation of verbs in lock+unlock pairs.
Previously --help output was too wide, it now fits in 80 columns.
Cosmetic changes in --help output only.
Co-developed-by: Claude Opus 4.7 <noreply@anthropic.com>
The order was random, different in the source code, different in the
verb list, different in --help. Order source code by --help to make
the next patch manageable.
Exposed in the dracut testsuite while adding tests for sysexts:
```
[ 2.972948] localhost (sd-merge)[510]: Validation of dm-verity signature failed via the kernel, trying userspace validation instead: Required key not available
[ 2.972993] localhost (sd-merge)[510]: Skipping file '/etc/verity.d/dracut.pem', suffix is not '.crt'.
[ 2.973045] localhost (sd-merge)[510]: No userspace dm-verity certificates found.
```
https://github.com/systemd/systemd/blob/658e5ac06f80ee2078b034f7cc483204d7f91c5e/src/shared/dissect-image.c#L3093
Exposed in the dracut testsuite while adding tests for sysexts:
```
[ 2.972948] localhost (sd-merge)[510]: Validation of dm-verity signature failed via the kernel, trying userspace validation instead: Required key not available
[ 2.972993] localhost (sd-merge)[510]: Skipping file '/etc/verity.d/dracut.pem', suffix is not '.crt'.
[ 2.973045] localhost (sd-merge)[510]: No userspace dm-verity certificates found.
```
This reverts commit 526f1594daec073269c3e70ee7914f6dd8740d5c.
This revert is necessary because the change breaks mDNS hostname stability
whenever a DNS-SD service calls UnregisterService. When a service
unregisters (e.g. on process restart), manager_refresh_rrs() clears and
re-adds all RRs in PROBING state, which sends a multicast announcement
(QR=1). The kernel reflects this back to resolved's own socket. Because
the local-address check was moved inside the query-only branch by the
reverted commit, the reply path in on_mdns_packet() is now unguarded.
The looped-back announcement matches the pending probe transaction and
completes it with DNS_TRANSACTION_SUCCESS. Since the zone item is still
in PROBING state (not ESTABLISHED), dns_zone_item_notify() sets
we_lost=true and calls dns_zone_item_conflict(), which invokes
manager_next_hostname() and renames the hostname (e.g. foo.local →
foo4.local). This happens reliably on every restart of any service using
RegisterService/UnregisterService (homebridge, avahi-compat wrappers,
etc.).
The top-level local-address check in on_mdns_packet() suppresses all
looped-back multicast traffic before the reply/query split. Restoring it
there is consistent with the overall design: dns_scope_check_conflicts()
already has its own manager_packet_from_local_address() guard and is
unaffected.
A more targeted long-term fix (e.g. guarding dns_transaction_process_reply()
for mDNS, or avoiding unnecessary re-probing of already-established records
in manager_refresh_rrs()) can be pursued separately.
These were only used to implicitly load libcryptsetup at startup.
dlopen_cryptsetup() now calls cryptsetup_enable_logging(NULL) itself,
and every code path that uses libcryptsetup calls dlopen_cryptsetup()
before doing so, so the upfront calls are no longer needed.
libfdisk is required by systemd-repart and it silently exits if dlopen fails
(unless the debug log level is set):
```
$ SYSTEMD_LOG_LEVEL=debug systemd-repart
Shared library 'libfdisk.so.1' is not available: libfdisk.so.1: cannot open shared object file: No such file or directory
$ echo $?
1
```
Follow-up for d49f3f287a
The verity signature partition content is a bare JSON object. Repart
pads it with zeros to fill the GPT partition. But when splitting out
the content as an individual file, the padding remains, so it's not
a valid text file.
jq started rejecting files with NUL bytes to fix a security issue:
https://github.com/jqlang/jq/commit/6374ae0bcdfe33a18eb0ae6db28493b1f34a0a5b
Trim the output when writing these files out.
If the passed in image path didn't end with .raw, we'd return an empty
string + suffix instead of the intended image + suffix path.
---
Also, fix two more nits that came up repeatedly in my searches.
is not used
add_veritysetup() is called unconditionally from add_root_mount() and
add_usr_mount() whenever in_initrd() is true, to generate units that
only activate if verity devices appear. However, when compiled without
libcryptsetup, this function returned a hard error, causing the entire
generator to fail even when no verity protection is in use.
Change the #else fallback to log a debug message and return 0, matching
the pattern already used by add_root_cryptsetup().
The .c file is shared between various fuzz-journal-* binaries. It
was added to 32bd43d768, but that is
somewhat ugly.
Let's add it to the alphabetially first fuzzer and share from there.
Follow-up for 32bd43d768 and
85b5acde86.
cryptsetup_set_minimal_pbkdf(), cryptsetup_get_token_as_json() and
cryptsetup_add_token_json() each take a struct crypt_device *cd, which
can only be obtained by first calling sym_crypt_init*() — and that
already requires dlopen_cryptsetup() to have succeeded. The internal
calls here were only implicitly re-loading a library the caller is
guaranteed to have already loaded.
These were only used to implicitly load libcryptsetup at startup.
dlopen_cryptsetup() now calls cryptsetup_enable_logging(NULL) itself,
and every code path that uses libcryptsetup calls dlopen_cryptsetup()
before doing so, so the upfront calls are no longer needed.
Convert systemd-cryptsetup, systemd-cryptenroll, systemd-veritysetup
and systemd-integritysetup to go through the existing dlopen wrapper
for libcryptsetup instead of linking the library directly. Each binary
calls dlopen_cryptsetup() at the start of its run() and uses the sym_*
variants for every libcryptsetup entry point.
Extend cryptsetup-util.{h,c} to cover the libcryptsetup symbols that
these binaries use and that the wrapper was missing:
crypt_activate_by_token_pin, crypt_deactivate, crypt_init_data_device,
crypt_keyslot_status, crypt_set_keyring_to_link (conditional on
HAVE_CRYPT_SET_KEYRING_TO_LINK), crypt_status and
crypt_token_external_path.
With no direct callers of crypt_free() left, drop the non-sym
crypt_freep cleanup variant and rename sym_crypt_freep back to
crypt_freep via DEFINE_TRIVIAL_CLEANUP_FUNC_FULL_RENAME, matching the
naming convention used by other dlopen wrappers (acl_freep,
xkb_context_unrefp, ...). Update the remaining users in src/shared,
src/repart, src/home and src/growfs to the new name.
The four affected meson targets switch from libcryptsetup to
libcryptsetup_cflags so they no longer record a DT_NEEDED entry for
libcryptsetup.so.12.
Convert the GnuTLS and libmicrohttpd usage in journal-remote to the
dlopen pattern used by other optional shared libraries. A new
src/shared/gnutls-util.{h,c} declares the GnuTLS entry points via
DLSYM_PROTOTYPE and resolves them in dlopen_gnutls(); microhttpd-util
is moved from src/journal-remote to src/shared and gains analogous
DLSYM_PROTOTYPEs plus dlopen_microhttpd(). Callers in journal-gatewayd,
journal-remote-main and microhttpd-util itself call the sym_* wrappers
and invoke dlopen_gnutls()/dlopen_microhttpd() at their entry points.
setup_gnutls_logger() no longer fails when libgnutls is missing at
runtime; it logs a notice and returns 0 so journal-gatewayd starts up
without TLS dependencies installed.
The meson files gain libgnutls_cflags and libmicrohttpd_cflags partial
dependencies that expose include paths and compile flags only. Every
systemd-journal-{gatewayd,remote,upload} target switches to the cflags
variant, dropping the direct libgnutls/libmicrohttpd link. The
gatewayd->remote object-sharing dance for microhttpd-util.o goes away
since the code now lives in libshared.
test-dlopen-so gains assertions for dlopen_gnutls and dlopen_microhttpd.
On Fedora Rawhide mount/umount is linked against libsystemd, which then
breaks the binaries in sanitizer runs, as we try to run instrumented
code from an uninstrumented binary:
bash-5.3# ldd /usr/bin/mount
linux-vdso.so.1 (0x00007fa757ef9000)
libmount.so.1 => /lib64/libmount.so.1 (0x00007fa757e84000)
libselinux.so.1 => /lib64/libselinux.so.1 (0x00007fa757e51000)
libc.so.6 => /lib64/libc.so.6 (0x00007fa757c56000)
libblkid.so.1 => /lib64/libblkid.so.1 (0x00007fa757c16000)
libsystemd.so.0 => /lib64/libsystemd.so.0 (0x00007fa757400000)
libpcre2-8.so.0 => /lib64/libpcre2-8.so.0 (0x00007fa75734f000)
/lib64/ld-linux-x86-64.so.2 (0x00007fa757efb000)
libclang_rt.asan.so => /usr/lib/clang/22/lib/x86_64-redhat-linux-gnu/libclang_rt.asan.so (0x00007fa756800000)
libm.so.6 => /lib64/libm.so.6 (0x00007fa7566e4000)
libgcc_s.so.1 => /lib64/libgcc_s.so.1 (0x00007fa7566b7000)
libstdc++.so.6 => /lib64/libstdc++.so.6 (0x00007fa756400000)
bash-5.3# mount
==458==ASan runtime does not come first in initial library list; you should either link runtime to your application or manually preload it with LD_PRELOAD.
This then breaks the whole machine, as mount is quite essential during
boot.
Let's just add mount/umount to the list of wrapped binaries to fix this.
Add --forward-journal=FILE|DIR to forward the container's journal
entries to the host via systemd-journal-remote. When specified,
nspawn starts systemd-journal-remote listening on a Unix socket,
bind-mounts it into the container at /run/host/journal/socket, and
passes a journal.forward_to_socket credential pointing to it.
Add --forward-journal-max-use=, --forward-journal-keep-free=,
--forward-journal-max-file-size=, and --forward-journal-max-files=
to configure disk usage limits for the forwarded journal.
Consolidate nspawn's per-machine on-disk state under a single runtime
directory at /run/systemd/nspawn/<machine>/. The container rootdir
mount point moves from /tmp/nspawn-root-XXXXXX to <runtime_dir>/root,
the unix-export directory moves from
/run/systemd/nspawn/unix-export/<machine> to <runtime_dir>/unix-export,
and the journal-remote socket lives at
<runtime_dir>/journal-remote-socket. Update ssh-generator and
ssh-proxy to follow the new unix-export path layout.
Extract fork_journal_remote() into fork-notify.{c,h} as a shared
helper used by both nspawn and vmspawn, replacing vmspawn's
start_systemd_journal_remote().
Extract runtime_directory_make() into path-lookup.{c,h} as a shared
helper used by both nspawn and vmspawn, replacing vmspawn's inline
runtime directory creation logic.
Co-developed-by: Claude Opus 4.6 <noreply@anthropic.com>
Add options to vmspawn to configure journal-remote disk usage limits
when forwarding journal entries from the VM. These are passed through
as --max-use=, --keep-free=, --max-file-size=, and --max-files=
command-line arguments to systemd-journal-remote.
Add --max-use=, --keep-free=, --max-file-size=, and --max-files=
command-line options to systemd-journal-remote to allow overriding the
corresponding settings from the configuration file.
Add $SYSTEMD_JOURNAL_REMOTE_CONFIG_FILE environment variable support
to systemd-journal-remote. When set, the specified file is used
instead of the default configuration file and drop-in directories.
When set to the empty string or /dev/null, configuration file parsing
is skipped entirely. vmspawn sets this to /dev/null in the child
process to avoid inheriting the host's journal-remote configuration.
Make fork_notify() argv parameter optional. When NULL is passed,
fork_notify() returns 0 in the child (with $NOTIFY_SOCKET set) and
lets the caller run custom code before exec. Returns 1 in the parent.
This allows vmspawn to set environment variables in the child without
polluting the parent process.
Co-developed-by: Claude Opus 4.6 <noreply@anthropic.com>
Replace the getopt_long()-based parser with the FOREACH_OPTION /
OPTION_* macros from src/shared/options.h, mirroring the recent
conversions of nspawn and vmspawn. Each option's metadata (long
name, short name, metavar and help text) now lives next to its
parsing logic, and the --help text is generated from those
definitions via option_parser_get_help_table() instead of being
hard-coded.
Positional file arguments are collected via
option_parser_get_args() rather than strv_skip(argv, optind).
Every dlopen_xxx() helper now takes an int log_level argument. It is
passed through to dlopen_many_sym_or_warn() (which in turn propagates it
to dlopen_verbose() for the library-not-installed case), and is used by
the fallback stub when support for the library is not compiled in to
emit a "<lib> support is not compiled in." message at the caller's level.
Callers pass LOG_DEBUG when gracefully degrading, or a higher level when
the failure should surface, and no longer need to log redundantly at the
call site.
As part of this, dlopen_bpf_full() (which already took a log_level) is
merged into dlopen_bpf() rather than keeping both.
The static inline fallbacks used to live in the headers, which required
pulling log.h in from every header that declared a dlopen_xxx(). Move
them into the .c files instead: the declaration is always outside the
#if HAVE_XXX block, the impl sits outside the outer #if HAVE_XXX wrap
with its own internal #if HAVE_XXX/#else/#endif, and apparmor-util.c,
idn-util.c, libmount-util.c and pam-util.c are now always compiled so
they can host their stubs.
jq started rejecting input that has NUL bytes to fix some security
issues,
so we need to trim the verity.sig json files, which are spat out with
the NUL bytes padding from the GPT partition content.
```
‣ Running postinstall script /home/runner/work/systemd/systemd/mkosi/mkosi.postinst.chroot…
jq: parse error: Invalid numeric literal at EOF at line 1, column 16384
‣ "/work/postinst final" returned non-zero exit code 5.
```
https://github.com/jqlang/jq/commit/6374ae0bcdfe33a18eb0ae6db28493b1f34a0a5bhttps://github.com/systemd/systemd/pull/41771 updates the producer.
It was bumped in a40d934007 but this
is hardly load bearing stuff so let's document the version we actually
require rather than the version that makes a hardly load bearing feature
work properly, especially since v2.41 is extremely new and requiring
distributions to have that is just unrealistic.
This doesn't actually change anything materially except documentation,
but it keeps us honest about depending on stuff from newer util-linux
because we happen to document reliance on an extremely new version.
FAIL: check-pointer-deref.cocci found issues in /home/runner/work/systemd/systemd/src/oom/:
diff -u -p /home/runner/work/systemd/systemd/src/oom/oomctl.c /tmp/nothing/oomctl.c
--- /home/runner/work/systemd/systemd/src/oom/oomctl.c
+++ /tmp/nothing/oomctl.c
@@ -107,7 +107,6 @@ static int parse_argv(int argc, char *ar
break;
}
- *ret_args = option_parser_get_args(&state);
return 1;
}
Coccinelle check(s) failed. For each flagged dereference, either:
- Add assert(param)/ASSERT_PTR(param) at the top of the function (if the parameter must not be NULL)
- Add an if (param) guard before the dereference (if NULL is valid)
- Add POINTER_MAY_BE_NULL(param) if NULL is okay for param
Follow-up for 143af68ee1
clang-tidy's misc-use-internal-linkage fires on BPF map declarations
(they have the SEC(".maps") attribute and must retain external linkage
so bpftool gen skeleton can resolve them as ELF symbols), and its
misc-include-cleaner flags errno.h as unused even where a /* IWYU
pragma: keep */ is present. clangd's own unused-includes analysis
emits the equivalent diagnostic independently of clang-tidy.
Add src/bpf/.clang-tidy and src/bpf/.clangd that inherit the parent
configs and scope these suppressions to BPF sources only.
compile_commands.json is generated by ninja from c_COMPILER rules, so
BPF programs (built via custom_target and thus emitted as CUSTOM_COMMAND
rules) never show up there. Clangd consequently falls back to
heuristics when opening .bpf.c files, with poor diagnostic fidelity.
Register a meson postconf script per BPF program that upserts an entry
into compile_commands.json using the same argv meson constructed for
the custom_target. The script runs after meson has written the DB,
substitutes @INPUT@/@OUTPUT@, and keys entries by source path so
repeated reconfigures don't accumulate duplicates.
The six .bpf.c files and their shared .bpf.h headers now live directly
under src/bpf/, rather than scattered across src/core/bpf/<prog>/,
src/network/bpf/<prog>/ and src/nsresourced/bpf/<prog>/.
All BPF compilation logic — the BPF_FRAMEWORK determination (clang/gcc/
bpftool/llvm-strip lookups), flag and command construction, vmlinux.h
handling, the bpf_programs list and the loop that builds the unstripped
object, the stripped object and the skeleton header for each program —
moves into src/bpf/meson.build. The top-level meson.build only keeps
option handling and the libbpf dependency. subdir('src/bpf') is pulled
up before config.h generation so that BPF_FRAMEWORK, HAVE_VMLINUX_H and
ENABLE_SYSCTL_BPF land in conf in time.
Skeleton wrapper headers (<prog>-skel.h) are now emitted by a
configure_file() template (src/bpf/bpf-skel-wrapper.h.in) at meson
setup, replacing the previously checked-in shims of the same name.
Consumer #include paths are flattened:
"bpf/<prog>/<prog>-skel.h" becomes "<prog>-skel.h",
"bpf/<prog>/<prog>-api.bpf.h" becomes "<prog>-api.bpf.h". src/bpf is
added to includes so the shared BPF headers resolve.
sysctl-write-event.h, now shared between userspace (networkd-sysctl.c)
and BPF (sysctl-monitor.bpf.c) from a single location, gains guarded
includes so pid_t and uint64_t resolve on both sides: vmlinux.h in the
BPF case (selected via __bpf__), stdint.h + sys/types.h otherwise.
fuzz-journald-util.c is compiled unconditionally even though fuzzing
tests aren't enabled. Only build it if fuzzing tests are configured.
This also ensure that the functions it uses from src/shared/tests.c are
available.
Fixes 32bd43d768Closes#39984
Signed-off-by: Chris Hofer <christian.hofer@codasip.com>
Both devices have -90 degrees mounted panels but they don't have the
quirk in kernel.
The Pocket 4 has been researched and it has an acpi accel matrix that
works when setting panel orientation at boot parameter.
The Pocket 3 hasn't been tested, but given it didn't had panel
orientation quirk is for sure that matrix is wrong for it.
Actually is pending the quirks for both devices in kernel but eventually
they will get merged. Till that happens is encourage that owners of
these devices set panel orientation boot parameter to right-up.
Fixes: #41036.
* 94af257c72 d/t/control: pull libmicrohttpd-dev for unit-tests suite
* 08263f18a4 d/t/control: pull libfdisk-dev for test suites
* e54175a0a4 Install new files for upstream build
jq started rejecting input that has NUL bytes to fix some security issues,
so we need to trim the verity.sig json files, which are spat out with
the NUL bytes padding from the GPT partition content.
‣ Running postinstall script /home/runner/work/systemd/systemd/mkosi/mkosi.postinst.chroot…
jq: parse error: Invalid numeric literal at EOF at line 1, column 16384
‣ "/work/postinst final" returned non-zero exit code 5.
https://github.com/jqlang/jq/commit/6374ae0bcdfe33a18eb0ae6db28493b1f34a0a5b
This is split out of #41543 but I think makes sense of its own.
It primary does one thing: ensure that initrds installed via the Linux EFI protocol are incremental in behaviour (i.e. we read the previously set initrd and combine it with ours). So far we'd simply not install any initrds at all in this case, which would break stuff.
THis is preparatory for #41543, but is generally the better, safer behaviour.
This also contains three minor changes which are purley prep work for #41543 but shouldn't hurt in the big picture.
Yu Watanabe
Zbigniew Jędrzejewski-Szmek
Christian Brauner
vlefebvre
Lennart Poettering
Antonio Alvarez Feijoo
Bret Comnes
Luca Boccassi
Nandakumar Raghavan
Daan De Meyer
Clayton Craft
Frantisek Sumsal
Chris Hofer
Kajus Naujokaitis
Heran Yang