In the CentOS Hyperscale SIG, we maintain a backport of systemd with
its own rpm spec forked from rawhide.
Let's make it easy to build upstream rpms using the Hyperscale spec
by adding a mkosi hyperscale profile.
In the CentOS Hyperscale SIG, we maintain a backport of systemd with
its own rpm spec forked from rawhide.
Let's make it easy to build upstream rpms using the Hyperscale spec
by adding a mkosi hyperscale profile.
Instead of including distribution specific files in the subimages,
let's have one common mkosi.pkgenv/ directory that handles all the
matching which is then included in the subimages.
This gives us more control on exactly how we do the matching.
* 10544812b3 Don't fix up vmlinuz locations if not required
* 9baf551923 Reduce amount of packages in default image
* 9e1a2f18b8 Add support for assert sections
* c7c6e2c0b1 ubuntu: Switch to devel as the default release
* 0822deb69c Improve logging when we can not extract kernel version from filename
* 1fd7ef3db7 Do not build default initrd if Initrds= is specified
* 611c8b46c8 Don't unconditionally sync when PackageCacheDirectory=/var
* ab37f24d38 README: add link to OBS docs for mkosi builds
* 02bf256ebb completion: add reuse ignore comments
* 10ccb0b04f Make sure not all subimages depend on default-initrd subimage
* 442e1ce0f1 mkosi-tools: systemd-boot-tools is now available for all arches
* 307fc1dba3 action: make it work when used as a submodule
* c37a55f91b config: serialize dataclass instances in our JSONEncoder
* f26cb34155 log: set terminal window title in complete_step while mkosi runs
* 280c78e681 Make sure inherited settings are applied for the default initrd
* ae4f2fd718 Fix typo
* 5644f3e83e build(deps): bump github/codeql-action from 3.29.7 to 3.30.5
* 7d7b26b8c9 build(deps): bump actions/checkout from 4.2.2 to 5.0.0
* e805253447 postmarketos: implement is_kernel_package
* dd51d2e019 postmarketos: provide missing i386-vars.fd
* e23e6de66b Do not relabel files when building extension image
* ab05ead5eb Bump various systemd version checks to 258
* 382cc8b450 preset: drop systemd-networkd-wait-online.service
* 477b6b2ed5 mkosi-initrd: install systemd-container in network profile
* 1d167c0b53 mkosi-initrd: add 89-ethernet.network for network profile
* b1e81dec3a mkosi-initrd: Install libseccomp explicitly
* 3c431a141c opensuse: repository non-oss-debug is invalid
* 62ab363149 ci: add an s390x build job for additional coverage
* 62765f2d07 mailmap: deduplicate Daan
* 47f963f40b Revert "ci: Disable PPC jobs temporarily"
* 6a48f83dda dnf: Always specify --best again
* 3702368616 sandbox: Make all relative paths absolute during argparsing
* a587af0bf7 dnf: Fix /var package cache directory check in package_subdirs()
* 56cdbc25b5 Always use repository metadata from /var package cache directory
* 669d4418a3 Add note on Encrypt=yes to Passphrase= docs
* caa129edae Drop BuildSourcesEphemeral=yes from default image config
* 7edca63478 Add devicetree-auto support for UKI
* 6cb1649074 Don't add ncdu for ppc64-le on Fedora
* e019d2d2a6 ci: Disable PPC jobs temporarily
Some of the functions were ignoring failure in cleanup, others weren't. If we
got a reply, it's better to use it, so ignore failures in cleanup everywhere.
This moves the open call earlier, so that we do any state-changing operations
if we actually managed to open the nonblocking fd. The code is easier to follow
this way and might be more robust.
Suprisingly, this fixes https://github.com/systemd/systemd/issues/39055: it
seems that run0 chowns /dev/stdin (in my case /dev/pts/0) to root:root, and the
second run0 can read and write stdin/stdout throught the already-open fds,
but fd_reopen fails.
Fixes https://github.com/systemd/systemd/issues/39055.
The enum definition only exists for cases like this, use it, instead of
its literal value.
This doesn't effectively change anything, but cleans up the code a bit.
```
==19541== 8 bytes in 1 blocks are still reachable in loss record 1 of 3
==19541== at 0x4841744: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==19541== by 0x50125C9: strdup (strdup.c:42)
==19541== by 0x4C55925: getusername_malloc (user-util.c:154)
==19541== by 0x1121D6: parse_argv_sudo_mode (run.c:1098)
==19541== by 0x123B13: run (run.c:3032)
==19541== by 0x124198: main (run.c:3100)
==19541==
==19541== 11 bytes in 1 blocks are still reachable in loss record 2 of 3
==19541== at 0x4841744: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==19541== by 0x50125C9: strdup (strdup.c:42)
==19541== by 0x112A9C: parse_argv_sudo_mode (run.c:1182)
==19541== by 0x123B13: run (run.c:3032)
==19541== by 0x124198: main (run.c:3100)
```
Instead of skipping libcore entirely when libmount is not available,
let's only compile out the pieces that need libmount. This makes the
meson logic much less complex and allows systemd-analyze to be built
when libmount is not available.
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.
Currently, if the libmount feature is disabled, we don't build libshared
and as a result skip building every other executable as well. Among
other things, this makes our nodeps CI builds kind of pointless since
hardly any code will be compiled.
Let's improve on the situation by making libmount properly optional in
libshared. Then, we only skip building the executables that actually
need libmount.
Currently, if the libmount feature is disabled, we don't build
libshared and as a result skip building every other executable as
well. Among other things, this makes our nodeps CI builds kind of
pointless since hardly any code will be compiled.
Let's improve on the situation by making libmount properly optional
in libshared. Then, we only skip building the executables that
actually need libmount.
running_in_chroot() will fail when a test is executed as a non-root
user without CAP_DAC_READ_SEARCH as it won't be able to access
/proc/1/root.
Let's make things more robust by skipping tests if we can't detect
if we're in a chroot or not, since if we can't even detect if we're
in a chroot or not, chances are we're missing the required privileges
to execute the test anyway.
Now that we dlopen() so many deps, it might happen by accident that we
end up dlopen()ening stuff when we entered a container, which we should
really avoid, to not mix host and container libraries.
Let's add a global variable we can set when we want to block dlopen() to
ever succeed. This is then checked primarily in
dlopen_many_sym_or_warn(), where we'll generate EPERM plus a log
message.
There are a couple of other places we invoke dlopen(), without going
through dlopen_many_sym_or_warn(). This adds the same check there.
The Beacn Mic's alt-mode behaves identically to it's primary mode from a
communication perspective, it just presents a different channel
configuration to ALSA.
boot_entry_parse_tries() replaces the id, which means the id
with the profile appended is lost (leaked) and replaced by a plain filename
in case there are tries suffixes. This means the wrong order is used in
displaying the entries in the menu, as the main profile is always last
given id_without_profile has the tries suffixes and sorts higher,
while the main profile has no id_without_profile and the id sorts lower
since it does not have the tries suffix.
Follow-up for 4301ad00ef
