The post-install depends on the -openrc files being present, and if the
dependency isn't explicitly specified then the device-* package may be
installed before the -openrc package, causing the post-install to fail.
MSM8916 has a single "modem" firmware for the audio DSP and modem.
In some early tests, the remote processor boots successfully at least
and the modem shows up in ofono. (Have not done further tests yet...)
However, audio stops working as soon as it is loaded
(because the audio DSP assumes control over the audio hardware).
For now let's just package it for all MSM8916 mainline devices
so it can be installed more easily. It is not used or installed
by default but it can be manually installed for testing.
At the moment, only "modem" is detected as rpmsg name.
On MSM8916 the audio DSP and modem are both on the same remote
processor ("hexagon").
Add it to the udev rule so the modem symlink is created on MSM8916.
This commit gets rid of our custom way of launching Plasma Mobile in
favor of using the upstream scripts. This is mostly in preparation for
the first commit after the one packaged here, as it'll move to a binary
launch script rather than using scripts.
Both sway and phosh require hardware acceleration (DRM) to work. Let's
activate that when we request one of them. Also don't let them fail
during startup on inputless setups (which is common on a raspberry pi).
Phosh is too slow to be unsable, but now, it'll at least be unusable out
of the box ;-)
Current default is performance, and that is for sure not a great choice
for battery. We can experiment with ondemand as well, but based on docs,
conservative is recommended.
[ci:skip-build]: build takes too long
Without this MR, elogind, and therefore weston will not start as it will
complain about cgroup mess:
[ 21.387150] elogind-daemon[1538]: Failed to determine whether /sys/fs/cgroup is a mount point: Symbolic link loop
[ 21.388341] elogind-daemon[1538]: Failed to mount cgroup at /sys/fs/cgroup/elogind: No such file or directory
Fix by applying this patch, thanks minlexx for pointing me to it! This
might possibly also be applicable to other kernel 3.4-based ports.
[ci:skip-build]: already built successfully in CI
On some devices a getty should run on the serial console.
Configure the getty by setting the deviceinfo_getty variable. The
format is "<port>;<baud rate>". For example, "ttyS0;115200".
A post-installation trigger in postmarketos-base checks
/etc/deviceinfo, and modifies /etc/inittab if the device should run
a getty.
This provides alsa usecase profiles for the audio routing in the
pinephone, pulseaudio also picks up this config.
[ci:skip-build]: already built successfully in CI
The framework check is supposed to make sure, that we always have the
same versions of multiple packages that belong to the same framework.
We must exclude git versions for this check to have it working properly.
The interesting question is of course: why was this working before?
The check is using the url field to figure out which packages belong to
a certain framework, and by chance, all packages that had a git version,
did also use single quotes around the URL:
url='http://qt-project.org/'
With the old APKINDEX parsing code, this was parsed as:
url: "'http://qt-project.org/'"
And therefore, this was considered a different URL than the URL with
double quotes (again, by chance!) used by the stable qt versions:
url: "http://qt-project.org/"
I have improved the APKBUILD parsing code in pmbootstrap today, and now
it is properly parsing both URLs.
Related: https://gitlab.com/postmarketOS/pmbootstrap/merge_requests/1837
