Building systemd with mkosi generally requires a very recent version
of systemd which might not be installed on the host. Let's configure
mkosi to look for extra executables in the build/ directory by default
so that we prefer systemd executables from the build directory over those
on the host as those on the host are likely to be too old.
Let's enable usage of a tools tree by default to simplify the setup
for new contributors and save them from having to install or upgrade
a bunch of extra tools to get mkosi working as expected.
We need packages from EPEL to be able to build CentOS Stream images
with a CentOS Stream tools tree so enable it. This is broken on CentOS
Stream 10 but given using a CentOS Stream tools tree is broken without
EPEL as well, we might as well enable it and just wait until the packages
are added to EPEL 10.
This directory is commonly used for virtual Python environments.
These are useful when developing to install different Python versions
as well as install tooling like mkosi and mypy in an isolated fashion
without influencing the global system.
Introduce the `systemd.break=` kernel command line option to allow
stopping the boot process at a certain point and spawn a debug shell.
After exiting this shell, the system will resume booting.
It accepts the following values:
- `pre-udev`: before starting to process kernel uevents (initrd and
host).
- `pre-basic`: before leaving early boot and regular services start
(initrd and host).
- `pre-mount`: before the root filesystem is mounted (initrd).
- `pre-switch-root`: before switching root (initrd).
The Azure MANA folks would like the PCI domain to be suppressed from
naming network interfaces. Let's introduce a somewhat generic way to do
this, without hardcoding anything to Azure.
Specifically: we'll ship a new hwdb entry that sets a new
ID_NET_NAME_INCLUDE_DOMAIN=0 property on relevant MANA devices. Then we
make net_id look for that property, and if it is set we simply suppress
the PCI domain.
(Untested as of now, needs feedback from Azure MANA folks that this
actually works and does what is requested here).
Move some config option in the right section, fixes the following warning:
```
mkosi.conf: Setting Credentials should be configured in [Runtime], not [Host].
mkosi.conf: Setting RuntimeBuildSources should be configured in [Runtime], not [Host].
mkosi.conf: Setting RuntimeScratch should be configured in [Runtime], not [Host].
mkosi.conf: Setting QemuSmp should be configured in [Runtime], not [Host].
mkosi.conf: Setting QemuSwtpm should be configured in [Runtime], not [Host].
mkosi.conf: Setting QemuVsock should be configured in [Runtime], not [Host].
mkosi.conf: Setting QemuKvm should be configured in [Runtime], not [Host].
```
Introduce the `systemd.break=` kernel command line option to allow stopping the
boot process at a certain point and spawn a debug shell. After exiting this
shell, the system will resume booting.
It accepts the following values:
- `pre-udev`: before starting to process kernel uevents (initrd and host).
- `pre-basic`: before leaving early boot and regular services start (initrd and
host).
- `pre-mount`: before the root filesystem is mounted (initrd).
- `pre-switch-root`: before switching root (initrd).
This generalizes and modernizes the code to acquire set of local caps,
based on the code for this in the condition logic. Uses PidRef, and
acquires the full quintuplet of caps.
This can be considered preparation to one day maybe build without
libcap.
If an enumerated device has ID_PROCESSING=1 property, and the service
manager does not know if the device has been processed by udevd
previously (that is, Device.deserialized_found does not have
DEVICE_FOUND_UDEV), then drop DEVICE_FOUND_UDEV flag from the device and
make the device not enter the active state.
Follow-up for 405be62f05, which was
reverted by c4fc22c4de.
Follow-up for 12807b5a49.
Otherwise, if a call of shovel() disabled the flags, the subsequent
calls do nothing even if there is something we need to read or write.
Fixes the following error:
```
Dec 19 02:19:39 run0[5618]: Error on PTY forwarding logic: Too many levels of symbolic links
```
This function pins the *API* FS, i.e. /proc/ + /sys/, not just any fs.
Hence clarify this in the name.
(At least we call these two fs "API (V)FS" in our codebase, hence
continue to do so here)
Previously, even if e.g. .rules files are unchanged, all .rules files
are reloaded when other kind of config files like .link files or
.hwdb.bin are changed, vice versa.
Then, save configurations by their source: udev.conf, command line
arguments, kernel command line options, and udev control.
Preparation to support reloading udev.conf in a later commit.
The man page says that --sign-kernel and --no-sign-kernel "override the
detection of whether to sign the Linux binary", so we should only
autodetect if neither are specified. But as of commit 02eabaffe9
("ukify: Add a unified interface for signing tools"), we autodetect even
when --no-sign-kernel is passed, which makes the flag useless.
The sign_kernel option is parsed using argparse.BooleanOptionalAction,
which sets it to either True, False, or None. commit 02eabaffe9
replaced `sign_kernel is None` with `not sign_kernel`. These are not the
same in Python, as the latter accepts False as well as None.
Restore the original check and fix type annotations accordingly.
Fixes: 02eabaffe9 ("ukify: Add a unified interface for signing tools")