This splits out common logic in keyutil and systemd-measure.
Note, previously d2i_PUBKEY_fp() was used, but now it is replaced with
d2i_PUBKEY(), as musl seems to not support reading from memstream.
The previous commit fixes all remaining violations of the check for
a very specific setup (Fedora Rawhide with as many dependencies as
possible installed). The linter job in CI runs that specific setup
so we enable the check in the clang-tidy config but do not yet enable
the clang-tidy test suite in meson by default as it will very likely
still fail in different setups than the one in CI.
- Do not use '%m' when errno is not set.
- Do not use pam_syslog_errno() when errno is not set.
- Use pam_debug_syslog_errno() rather than log_debug_errno().
- Use 0 rather than PAM_SUCCESS in the function that returns negative
errno, though PAM_SUCCESS == 0, hence that does not change any
behavior.
- Append missing full stop in the log message.
Currently, the result of pam_debug_syslog_errno() is unused, hence this
does not change anything and may be slightly redundant. But let's follow
our usual coding style.
- pam_log_oom() passes ENOMEM rather than -ENOMEM, hence previously
pam_log_oom() did not return PAM_BUF_ERR.
- We may (mistakenly) pass SYNTHETIC_ERRNO(). Let's gracefully handle that.
- Introduce errno_to_pam_error() helper function.
Currently, when a socket unit specifies SmackLabel=,
the label is not applied to the underlying Unix socket file or its file
descriptor.
This change ensures that the SMACK label is applied both to the
Unix socket path on the filesystem and to all associated socket FDs
when the socket is created.
Testing:
- Tested on Fedora 43 with kernel 6.17.7 with SMACK enabled.
- Created a systemd socket unit:
[Unit]
Description=UNIX socket
[Socket]
ListenStream=/run/test.sock
SmackLabel=label
[Install]
WantedBy=sockets.target
- Created a corresponding service:
[Unit]
Description=UNIX socket service
Requires=test.socket
[Service]
ExecStart=/usr/bin/socat -v - -
StandardInput=socket
[Install]
WantedBy=multi-user.target
- Verified SMACK labels using getfattr:
- Without SmackLabel:
```
# file: run/test.sock
security.SMACK64="_"
# file: proc/<pid>/fd/*
security.SMACK64="*"
```
- With SmackLabel=label:
```
# file: run/test.sock
security.SMACK64="label"
# file: proc/<pid>/fd/*
security.SMACK64="label"
```
The function socket_address_listen() is declared at shared/socket-label.c,
however its prototype was in basic/socket-util.h. This moves the
prototype to shared/socket-label.h.
Also, enum SocketAddressBindIPv6Only is not used anymore in basic/*.[ch].
Let's move the definition and its string table to shared/socket-label.[ch].
Follow-up for b25a930f0e.
When inheriting file descriptors for stdout/stderr (either from stdin or
when making stderr inherit from stdout), we previously just assumed they
would be writable and dup'd them. This could lead to broken setups if
the inherited FD was actually opened read-only.
Before dup'ing any inherited FDs to stdout/stderr, verify they are
actually writable using the new fd_is_writable() helper. If not, fall
back to /dev/null (or reopen the terminal in the TTY case) with a
warning, rather than silently creating a broken setup where output
operations would fail.
This reverts commit 3ae7d8fd87.
Now utmp support is always disabled when building with musl,
and all definitions are unused in that case. Let's remove it.
This reverts commit bf9bc5beb0.
libutmps does not support utmpxname(), the function always fails
with ENOSYS, and always uses their own file.
However, our code relies on the funtion needs to succeed.
Let's revert the change now, and revisit later when musl users
request to support libutmps.
Nowadays, we append an extra NUL for each data if possible for safety.
We already do the same for example at write_to_kmsg(), log_do_context(),
write_to_journal(), log_struct_iovec_internal(), and so on.
This does not change any behavior, as the iov_len field is unchanged.
Now the controller part is always ignored, hence let's skip check for
the controller part of the spec. This also make it acceppt unnormalized
path. Previously paths were checked by path_is_normalized(), but now
checked by path_is_safe(). Also, now this mapps an empty path to NULL.
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().
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().
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.
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
Reverts systemd/systemd#38680
After taking a closer look I'm not convinced by the approach, see below.
First of all, all other SD_PATH_SEARCH_* are either somewhat generic,
i.e. encode the common prefix for configurations, binaries, etc., or are
subdirectories under systemd/ hence in our own "domain". The
tmpfiles/sysctl/binfmt we don't prefix with "systemd" precisely because
the concept is generic and there're actually other impls of them. A
specific SD_PATH_SEARCH_SYSCTL doesn't fit into our existing scheme.
Instead something along the lines of "SEARCH_SYSTEM_CONFIGURATION" shall
be introduced, and consumers will just suffix
sysctl.d/tmpfiles.d/binfmt.d for the final result.
And secondly, I don't grok why systemd-sysctl now unnecessarily calls
into sd-path to obtain the fixed search path. None of our other tools do
that.
-----------
An alternate approach, SD_PATH_SYSTEM_SEARCH_CONFIGURATION, which does
exactly above, will be introduced instead. It provides a universal
interface for querying any system config with our idiomatic
/etc/:/run/:/usr/local/lib/:/usr/lib/ hierarchy.
Its value depends on how strerror_r() implemented, and the
implementations of the function in glibc and musl are actually
different. Let's drop the checks.
musl provides XSI compliant strerror_r(), and it is slightly different
from the one by glibc.
Let's introduce a tiny wrapper to convert XSI strerror_r() to GNU one.
The wrapper also patches musl's spurious catchall error message.
glibc's sys/stat.h includes linux/stat.h, and we have copy of it from
the latest kernel, hence all new flags are always defined.
However, musl's sys/stat.h does not include linux/stat.h, and moreover,
they conflict with each other, hence we cannot include both header
simultaneously. Let's define missing macros to support musl.
Yu Watanabe