It is a good practice to install files relative to our own variables
https://www.bassi.io/articles/2018/03/15/pkg-config-and-paths/
and it is required on systems like NixOS.
Unfortunately, systemd allows overriding the root prefix,
see also 1c2c7c6cb3,
so we cannot just do that.
Let's at least make the systemd installation prefix overridable.
This will also allow us to drop `systemdsystemunitdir` option since
systemd hardcodes it to `${rootprefix}/lib/systemd/system`.
Since bash-completion 2.9, it was no longer possible to override
the completionsdir through prefix. [1] In 2.10, the overridability
was re-estabilished but this time through datadir variable. [2]
This should not really matter except for developers installing the project
into a custom prefix or distros using per-package prefixes like NixOS.
[1]: 81ba2c7e7d
[2]: https://github.com/scop/bash-completion/pull/344
The README says: "A test suite that can be used to interact with a fake device is installed when configured with `-Ddaemon=true` and `-Dtests=true`", so actually only install these tests when tests are enabled.
From [Lennart’s answer on systemd-devel][1]:
> fwupd.target should not list the various network management solutions,
> that is unnecessary, and ordering after network.target should be
> sufficient to achieve the same, in a generic fashion. i.e. network
> managers should order themselves before network.target, so that
> ordering yourself after that automatically orders you after all
> implementations at once...
So, remove the unneeded targets. As the service unit is not
installed/enabled, and instead called by a timer, the system will most
probably be online already.
[1]: https://lists.freedesktop.org/archives/systemd-devel/2020-March/044205.html
"Best practices for starting unit only if network is online"
A Jcat file can be used to store GPG, PKCS-7 and SHA-256 checksums for multiple
files. This allows us to sign a firmware or metadata multiple times (perhaps
by the OEM and also then the LVFS) which further decentralizes the trust model
of the LVFS.
The Jcat format was chosen as the Microsoft catalog format is nonfree and not
documented. We also don't want to modify an existing .cat file created from WU
as this may make it unsuitable to use on Windows.
More information can be found here: https://github.com/hughsie/libjcat
If the measurements are missing but it's a UEFI system, it's a good indication
that the user has secure boot turned off.
Notify the user on the UEFI device through a non-fatal `UpdateMessage`
To accomplish this, move fu-uefi-vars into the plugin library for other plugins to use
This allows it to be refreshed anytime the daemon updates rather
than once a day by a systemd job.
As part of this change, remove the logging from `fwupdmgr` which
was only used for motd purposes.
Some hardware does not handle upgrading from version 1.2.2 to 1.2.4 and instead
needs to be upgraded from 1.2.2->1.2.3->1.2.4 so that on-device metadata can be
migrated correctly.
Add a new per-device flag `install-all-releases` which causes the daemon to not
skip directly to the newest release. This is designed to be set from a quirk
file.
This can obviously only be used for devices that can apply firmware "live" and
thus do not need a reboot or system shutdown to actually apply the firmware.
This also needs the cabinet archive to ship multiple versions of the firmware,
and for the metainfo.xml file to refer to multiple release objects.
In some CI infrastructures other devices show up and so the webcam
is not automatically picked.
```
Enabling fwupd-tests remote...
Successfully enabled remote
Update the device hash database...
Choose a device:
0. Cancel
1. 08d460be0f1f9f128413f816022a6439e0078018 (Integrated Webcam™)
2. bbbf1ce3d1cf15550c3760b354592040292415bb (Virtio network device)
Request canceled
FAIL: fwupd/fwupdmgr.test (Child process exited with code 2)
SUMMARY: total=1; passed=0; skipped=0; failed=1; user=0.0s; system=0.0s; maxrss=11736
FAIL: fwupd/fwupdmgr.test (Child process exited with code 2)
autopkgtest [22:09:32]: test ci: -----------------------]
ci FAIL non-zero exit status 2
autopkgtest [22:09:33]: test ci: - - - - - - - - - - results - - - - - - - - - -
autopkgtest [22:09:33]: @@@@@@@@@@@@@@@@@@@@ summary
ci FAIL non-zero exit status 2
```
To debug flashing failures it's sometimes requried to get a SPI dump of the
hardware to analysis.
Add a debug-only command that lets us dump the device from the engine.
During startup we do 1898 persistent allocations to load the quirk files, which
equates to ~90kb of RSS. Use libxmlb to create a mmap'able store we can query
with XPath queries at runtime.
Makes `fwupd-refresh.service` strictly opt-in.
Some distros are defaulting to all systemd services on and causing
more refreshes than desirable by default, especially when using
both `gnome-software` and `fwupd-refresh.service`
It turns out there is some bug in systemd v242 or less that runtime
directories can't be used. So only populate motd when we know that
we have a newer systemd
`fwupd-refresh.service` uses `DynamicUser=true` which causes systemd
to make `/var/cache/fwupd` a symlink to `/var/cache/private/fwupd`.
Individual units aren't allowed to access this directory, only the ones
with the directive. This means that `fwupd.service` stops working as
soon as a user tries to start `fwupd-refresh.service`.
The bug details are present in
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=941360
This allows us to easily build just libfwupd in a flatpak manifest without
installing dozens of deps to build things we're just going to delete anyway.
Mostly for consistency purpose. Details:
* It's confusing that internally the functions for `FwupdClient` use
`upgrade` in the name.
* The logical antonym of `downgrade` is `upgrade` not `update`
* People who don't use the tool frequently may try `get-upgrades`
Fixes Debian bug https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=921820
Introduce a new --log option to fwupdmgr that will log stdout to an argument.
If run under systemd, prefix that argument with $RUNTIME_DIRECTORY.
Add a new systemd unit and associated timer to regularly refresh metadata.
After the metadata refresh is complete, save the output to the motd location.
The timer and service are disabled by default and can be enabled by an admin.
In many plugins we've wanted to use ->prepare_firmware() to parse the firmware
ahead of ->detach() and ->write_firmware() but this has the limitation that it
can only return a single blob of data.
For many devices, multiple binary blobs are required from one parsed image,
for instance providing signatures, config and data blobs that have to be pushed
to the device in different way.
This also means we parse the firmware *before* we ask the user to detach.
Break the internal FuDevice API to support these firmware types as they become
more popular.
This also allows us to move the Intel HEX and SREC parsing out of the dfu plugin
as they are used by a few plugins now, and resolving symbols between plugins
isn't exactly awesome.
This allows several things, for instance:
* Adding or removing blacklisted plugins or devices
* Changing the idle timeout where allowed
...without a user needing to manually modify a configuration file.
This information was a predecessor to metadata provided by LVFS with
actual files associated. It's not useful to 99% of the machines it runs
on, and future VIA metadata should come directly with releases on LVFS.
The offline updates environment is special, and we have to be careful to delete
the trigger before doing anything that can fail to avoid boot loops.
For this reason, split it out to a simple self-contained binary that is easy to
understand.
If another offline update task is run at the same time, e.g. pk-offline-update
from PackageKit then we might corrupt the package database when the client
D-Bus request times out.
Copy the fixes from PackageKit so that the offline updates work together.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1685471
This currently just outputs the current list of devices with releases and makes
it possible to integrate firmware version reporting with other tools like mgmt.
