Following a kernel upgrade, HID_GENERIC was changed from =y to =m,
however it was not included in the initramfs, causing the keyboard to
not respond and thus the FDE unlocking to be broken. Fix this by adding
it to the initramfs so the keyboard works again.
Signed-off-by: Ferass El Hafidi <funderscore@postmarketos.org>
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/8064>
Changed in my personal tree (sdm710-mainline) have been merged into
sdm670-mainline common tree as those SoCs are very similar (essentially
the same).
Currently trying to get ADSP working, so this patch also adds config
files for Hexagon DSP.
Since I got U-Boot working on this device, I also added systemd-boot.
This way I can provision rootfs on MicroSD and dual-boot into Android
for grabbing more logs, traces and firmware files.
Signed-off-by: Alicja Michalska <ellyqw@mainlining.org>
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/8021>
This fixes a race condition between the panel initialization and the console.
I cannot reproduce it myself, but others report that this change fixes the
race condition for them, so add it anyways.
Signed-off-by: Aster Boese <asterboese@mailbox.org>
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7953>
Moved them to a seperate repository with meson as buildsystem. That
makes it much more maintainable in a single repository, suitable for
upstreaming in Alpine, allows other distros to use it, and cleans up
A LOT of tech dept.
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7677>
Apparently starting after MM isn't always enough time for the location
service to be up, so this adds a periodic restart of the systemd service
every 30 seconds for up to 5 minutes.
Signed-off-by: Clayton Craft <craftyguy@postmarketos.org>
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/8041>
On devices using a combined initramfs, init_2nd.sh is exec'd directly
from the first stage without ever calling load_modules(). This means the
loop kernel module (CONFIG_BLK_DEV_LOOP=m) is not guaranteed to be
loaded before mount_subpartitions() runs, causing losetup to fail:
losetup: /dev/sda13: failed to set up loop device: No such file or directory
ERROR: failed to mount subpartitions!
On the split initramfs path (init.sh), load_modules is called explicitly
before jumping to init_2nd.sh, so this regression did not affect those
devices.
Fix by explicitly loading the loop module in mount_subpartitions(),
right before the loop that searches for partitions. This location was
suggested by Clayton Craft.
Regression introduced by: 86bdb82116 (use loop device for subpartitions)
Tested on: oneplus-cheeseburger (combined initramfs, msm8998, loop=m)
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/8034>
[Warning!] deviceinfo_append_dtb: Property is set to its default value (false) and can be removed.
[Warning!] deviceinfo_append_dtb_lts: Property is set to its default value (false) and can be removed.
[Warning!] deviceinfo_append_dtb_exynos4: Property is set to its default value (false) and can be removed.
sed -i '/^deviceinfo_append_dtb\(_.*\)\?="false"/d' device/*/*/deviceinfo
[Warning!] deviceinfo_external_storage: Property is set to its default value (false) and can be removed.
sed -i '/^deviceinfo_external_storage="false"/d' device/*/*/deviceinfo
[Warning!] deviceinfo_disable_dhcpd: Property is set to its default value (false) and can be removed.
sed -i '/^deviceinfo_disable_dhcpd="false"/d' device/*/*/deviceinfo
[Warning!] deviceinfo_bootimg_qcdt: Property is set to its default value (false) and can be removed.
sed -i '/^deviceinfo_bootimg_qcdt="false"/d' device/*/*/deviceinfo
Part-of: <https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/8035>
