Before:
```
Benchmark #1: ./skipcpio/skipcpio test-5.10.15-200.fc33.x86_64.img >/dev/null
Time (mean ± σ): 125.5 ms ± 0.9 ms [User: 97.4 ms, System: 27.9 ms]
Range (min … max): 124.8 ms … 129.4 ms 23 runs
```
After:
```
Benchmark #1: ./skipcpio/skipcpio test-5.10.15-200.fc33.x86_64.img >/dev/null
Time (mean ± σ): 12.2 ms ± 0.3 ms [User: 2.7 ms, System: 9.5 ms]
Range (min … max): 11.7 ms … 13.6 ms 212 runs
```
Besides the speedup, skipcpio now parses the cpio header and is not
falsely ending when the early cpio payload contains `TRAILER!!!`.
Fixes: https://github.com/dracutdevs/dracut/issues/1123
Adding initial packit [1] support for fedora maintenance which
should automate the release process from upstream release to fedora.
1. https://packit.dev
There is no `fstab_lines_l` variable used before.
The `--include` option was turned into `++include` and therefore the
switch option is not reachable anymore.
If SIGWHATEVER will be processed after fsfreeze -f, but before fsfreeze
-u we will end up with /boot/ never unfrozen, let's try to minimize risk of this.
on my system (ThinkPad X201s) I have several modalias entries that are
empty:
/sys/bus/platform/devices/dock.0/modalias
/sys/bus/platform/devices/dock.1/modalias
/sys/bus/platform/devices/dock.2/modalias
executing the drm module setup thus results in the following errors
dracut-install: No SOURCE argument given
when calling dracut_instmods and hostonly mode is enabled.
Skip those entries, as there are no modules to load in this case
anyways.
Note: it is not sufficient to use [['s -s test (file size is greater
than zero), as all those files are 4096B if you stat them.
In the current state, services that depend on network need to
use dracut hooks, since nothing with pull in the network
targets into the transaction.
In the future, it would be nice to provide developers on systemd-only
systems the possibility to not use dracut hooks at all, but simply put
normal systemd services into the initrd.
Also, some modules even right now depend on systemd ordering, like
cryptsetup, so let's make sure, that the ordering inside systemd work
properly as well.
Instead of running directly NM with --debug, create a drop-in in
nm-config.sh. This will make sure, that we can get the debug output
regardless the way how NM will be started.
The current name of this bool is kinda stupid. Based on the manpage
setting it to 0 turns off fcoe, which means that nofcoe=1 should mean
that it is on.
Let's just do the same thing as with rd.lvm=0, rd.luks=0,....
A test of the form
if ! command; then
_ret=$?
...
return _ret
fi
does not capture the return code of `command`, but the negation of the
return code, leaving _ret == 0. The test of this form in inst_multiple
has been refactored to capture and return the right value.
Invalid space broke the creation of the keyboard subdirectories,
leading to a symbolic link
usr/share/consolefonts,consoletrans,keymaps,unimaps ->
/usr/lib/kbd/consolefonts,consoletrans,keymaps,unimaps
in the created initramfs.
With this fix the correct symlinks are created again:
usr/share/consolefonts -> /usr/lib/kbd/consolefonts
usr/share/consoletrans -> /usr/lib/kbd/consoletrans
usr/share/keymaps -> /usr/lib/kbd/keymaps
usr/share/unimaps -> /usr/lib/kbd/unimaps
Because some of the CI tests fail randomly while grepping for the
test success marker, let's be specific of the file format grep will
search to eleminate all failure sources.
Creating the /docs directory which will contain all project related
documentation and will also host the projects github based website
in the future.
Adding RELEASE.md which is a document that outlines the release process.
Harald Hoyer