- kconfig rebased onto jhovold's defconfig (many things are now modules,
but also includes initial camera support and other things)
- set CONFIG_FRAMEBUFFER_CONSOLE_DEFERRED_TAKEOVER=y, to try and hide
some super early kernel messages printed on console
- enabled kconfig options for supporting USB tethering e.g. w/ android
[ci:skip-build]: already built successfully in CI
With the new changes in lk2nd that allow us to support extlinux.conf,
we can now generate image that would be generic to these devices.
[ci:skip-build]: already built successfully in CI
Tested on:
* odroid-xu4
* samsung-klimtlte
* samsung-manta
Only regression I spotted is that usb networking no longer works on
samsung-klimtlte and samsung-manta (but wifi works fine, so still
possible to login to the device).
The boot partition on the device is 8.0 MB, while built boot.img is
8.8 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
[ci:skip-build] already built successfully in CI
The boot partition on the device is 8.0 MB, while built boot.img
is 8.7 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
Based on device PIT the boot partition is 28 MB, while built boot.img
is 31 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
Based on device PIT the boot partition is 28 MB, while built boot.img
is 31 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
.. when using mainline kernel. The boot partition on the device is
only 8 MB, while built boot.img is 11 MB. Fix by decreasing initramfs
size by depending on postmarketos-initramfs-minimal. Problems were
likely introduced by postmarketOS/pmaports!5000.
.. when using mainline kernel. The boot partition on the device is
only 8 MB, while built boot.img is 11 MB. Fix by decreasing initramfs
size by depending on postmarketos-initramfs-minimal. Problems were
likely introduced by postmarketOS/pmaports!5000.
Based on device PIT the boot partition is 40 MB, while built boot.img
is 41 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
Based on device PIT the boot partition is 40 MB, while built boot.img
is 41 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
Based on device PIT the boot partition is 10 MB, while built boot.img
is 12 MB. Fix by decreasing initramfs size by depending on
postmarketos-initramfs-minimal. Problems were likely introduced by
postmarketOS/pmaports!5000.
- Enable XFS (requested by @ell1e)
- Disable CONFIG_RT_GROUP_SCHED (kconfigcheck-community requirement)
[ci:skip-build] already built successfully in CI
This new daemon implements the same D-Bus API as callaudiod
in aports community/callaudiod, but works with the new audio profiles
generated by PulseAudio v17 and PipeWire v1.2 for pine64-pinephone
that the aports package's daemon does not.
Installs that use pipewire-pulse for audio (eg SXMO) will automatically start
having working call audio with this update. Installs that use pulseaudio
have to wait for pulseaudio's crash on startup to be fixed first
( https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/68586 ).
Ref: #2968
[ci:skip-build] already built successfully in CI
I use the PineTab 2 for various tasks on a semi-regular basis, so I
would like to co-maintain it.
---
[ci:skip-vercheck]:
Since this doesn't affect the resulting package as Co-Maintainer is a
postmarketOS schema extension that abuild doesn't read, there is no need
to bump pkgrel for this.
[ci:skip-build]: already built successfully in CI
- Added some modules to fix booting from emmc/nvme (fixes#2953)
- Partially fixes debug shell (#2955): the OSK still shows partially,
but at least now there's some usable console/shell with an external
usb keyboard.
[ci:skip-build] already built successfully in CI
Since the v6.9 upgrade the kernel supports Type-C functionality for
Fairphone 4, but I forgot to enable the relevant kconfig options then.
Enable them now along with a bunch of options related to USB that come
in useful. Also prepare some options that will be used in the future
with DisplayPort over USB-C (e.g. SBU mux).
[ci:skip-build]: already built successfully in CI
... and install the firmware package using _pmb_recommends instead.
Applied to both downstream and mainline ports.
[ci:skip-build]: already built successfully in CI
With the latest kernel bump the config was updated based on the output
of:
make ARCH="$_carch" CC="${CC:-gcc}" defconfig KBUILD_DEFCONFIG=librem5_defconfig
but there are other changes that pop up when running
pmbootstrap kconfig edit linux-purism-librem5
This commit updates the config so that future config empty edits don't
generate any changes.
This is still pretty in sync with the Purism config, besides passing the
`kconfig check` and disablling SELinux as that causes boot failures.
Signed-off-by: Alistair Francis <alistair@alistair23.me>
[ci:skip-build]: already built successfully in CI
Updated to latest collabora cros kernel. Fixes audio on mt8192 devices
and fixes display issues on all devices using this kernel.
[ci:skip-build]: already built successfully in CI
Add the required dependencies and config for both hexagonrpcd and
libssc so that sensors start working.
[ci:skip-build]: already built successfully in CI
Add the required dependencies and config for both hexagonrpcd and
libssc so that sensors start working.
[ci:skip-build]: already built successfully in CI
The device already loads plenty of nonfree-firmware into the
co-processors during the bootloader stages, and also once booted in
Linux won't really be useful without everything.
Additionally I doubt anyone tests that case, so just require nonfree
firmware unconditionally to simplify maintenance.
This gets Bluetooth working for the Sparklan card.
btattach doesn't work on the Librem5 at the moment (see [1]) so even
though it's deprecated let's use hciattach to get things working.
1: https://source.puri.sm/Librem5/librem5-base/-/issues/76
Signed-off-by: Alistair Francis <alistair@alistair23.me>
[ci:skip-build] already built successfully in CI
I made an oversight in pmaports!5295 which caused the post-install
script to try enabling the sd8x-wlan service even if the downstream
subpackage isn't installed. This causes errors when creating a new
mainline rootfs, so fix it by moving the install variable declaration to
the subpackage function.
Patches were rebased, sadly nothing was dropped :(
All kconfig changes are from simply running `kconfig migrate` and
accepting all defaults when prompted
[ci:skip-build]: already built successfully in CI
Resolves this harmless, but annoying error on mainline:
* Enabling wlan driver ...
/lib/rc/sh/openrc-run.sh: line 14: can't create /sys/devices/platform/sd8x-rfkill/pwr_ctrl: nonexistent directory
* ERROR: sd8x-wlan failed to start
[ci:skip-build]: already built successfully in CI
This also increases the (necessarily) hardcoded initrd end by 16MiB in
the DTS, which is necessary to make the phone boot with the new
initramfs (and in general any initramfs which exceeds 2624KiB).
This seems to be some Samsung's proprietary verified boot solution.
After a recent update, this started bootlooping my phone.
I have no idea why it only fired now, but pmOS does not use this
feature anyway, so it won't hurt to disable it.
[ci:skip-build] Already built fine once
Without it some devices will fail to mount the boot partition:
FAT-fs (mmcblk1p1): IO charset ascii not found
Also disable CONFIG_RT_GROUP_SCHED.
Reported-by: Anton Bambura <jenneron@protonmail.com>
Signed-off-by: Ferass El Hafidi <funderscore@postmarketos.org>
[ci:skip-build]: already built successfully in CI
This cmdline argument is now a no-op. Drop it everywhere and add a new
CI check to enforce this.
Adjust the deprecated "minimal" initramfs variant to use
pmos.debug-shell to enable logging instead. It doesn't /support/
dropping to a shell, so the variable takes on a different meaning. But
that's ok since it's at least more consistent.
[ci:ignore-count]
[ci:skip-vercheck]
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Tweaked-by: Oliver Smith <ollieparanoid@postmarketos.org>
[ci:skip-build]: kernel builds are very likely to go through now
* drop maxcpus=0, which was added with 73a033328e ("device-nokia-n900: set
uniprocessor mode for kernel (MR 4299)") while troubleshooting. a better
fix was eventually implemented with 3e737241ba ("linux-postmarketos-omap:
N900: disable twl off idle features (MR 4317)")
* drop PMOS_NO_OUTPUT_REDIRECT, which was obsoleted by 7c98127746
("postmarketos-initramfs: switch to udev and kmod (MR 5000)")
Use stock WiFi firmware instead of the generic linux-firmware-brcm
firmware. The stock firmware is newer and seems to be better suited
for the BCM4334B2 variant used in these tablets.
This fixes timeout errors from the brcmfmac driver and seems to
slightly improve download/upload speeds (by about 1MB/s).
trailblazer is a bleeding edge, unstable target for EFI capable ARM64
devices. It runs linux-next and should otherwise be adjusted to support
as many devices as possible.
The goal of trailblazer is to motivate more upstream kernel development by
providing a tighter feedback loop (patches land in -next weeks before they
make it into mainline or stable), as well as encouraging more generic
solutions to the device specific hacks we have in userspace.
On supported devices, trailblazer should reflect the current state of
upstream. It allows users to easily appreciate what feature are
unavailable, and more importantly appreciate when a new feature does get
added upstream.
Due to the instability of linux-next, the plan is to have BPO build
trailblazer images nightly, so if next breaks then the images should
only be broken for a day.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Introduce a linux-next kernel with postmarketOS configs and some
additional device specific configs enabled.
This kernel package will be updated daily to the latest -next tag.
breakages expected. It is designed to be used in tandem with a generic
ARM64 device package to provide a bleeding edge option for those with
EFI capable bootloaders.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
I'm not able to give this device the attention it deserves.
This also removes me from maintaining fw and kernel packages used by the
L5, and adds the current L5 maintainers to those packages.
[ci:skip-vercheck]
[ci:skip-build]: already built successfully in CI
Depend on msm-firmware-loader, firmware- and soc- packages.
Sort deps alphabetically.
Tidy up deviceinfo:
* add flash_kernel_on_update
* remove unneded things from kernel cmdline
[ci:skip-build]: already built successfully in CI
Currently it requries the check_code result check in the
vendor uboot to be disabled as explained in the wiki.
Booting directly via reconfigured boot order still
does not work. It hangs somewhere in the kernel.
Firmware for Wifi/BT is not included yet.
It could use the linux-postmarketos-omap kernel in the future,
but there are still some dirty things in there e.g. for sound
which are not suitable for a multi-device kernel.
[ci:skip-build]: already built successfully in CI
Bump to the 6.6.32 release, including updating the config with the
latest changes from Purism.
Signed-off-by: Alistair Francis <alistair@alistair23.me>
[ci:skip-build]: already built successfully in CI
PineTab 2 is a successor to the original PineTab from PINE64 released
in mid-2023.
The tablet came with DanctNIX's fork of Arch Linux ARM, comes with
K Desktop Environment as the UI and available with 64GB eMMC/4GB RAM
or 128GB eMMC/8GB RAM.
Changelog since 6.7:
General:
* Adreno 5xx preemtion is now disabled due to hangs it causes.
This makes GPU a lot more cooperative on these platforms.
* Fixed absence of zap-shader region which caused devices with
GPU enabled to freeze the whole system.
* Fixed ath10k Wi-Fi not connecting to 5Ghz networks.
* Incorporated in-kernel pd-mapper patchset, therefore userspace
pd-mapper daemon is not required.
* Kernel cmdline from bootloader is no longer ignored completely,
in other words, not hardcoded. You can now modify cmdline in
deviceinfo file and changes in that file will not be ignored.
Devices:
* Xiaomi Redmi Note 7 (lavender):
* split device trees into 2 variants by panel type (Tianma & BOE)
* GPU + panel support for 2 variants
* Vibrator support
* Connectivity: modem, Wi-Fi and bluetooth support.
* Other/all devices:
* update device trees according to upstream USB stack changes
(related to USB 3 phy support)
Install script into /etc/profile.d with workaround for mesa bugs.
This makes most SDM660 devices usable for general people out of the box.
Signed-off-by: Alexey Minnekhanov <alexeymin@postmarketos.org>
Add msm-firmware-loader, soc-qcom-sdm660-rproc, firmware-xiaomi-lavender
and firmware-qcom-adreno-a530 as dependency. This is necessary to make
Wi-Fi and modem work.
Add support for panel drivers.
commit 933f2b5703 ("device-nokia-n900: improve sxmo-dwm support [MR 5227]")
accidentally removed sxmo lock state initialization. Restore the lock state
initialization
Phoc now defaults to 1,25 as scale on skomer, which while not perfect,
is good enough and lets the device be unlocked. As such, let's remove
this configuration file and encourage improvements upstream instead.
See https://gitlab.com/postmarketOS/pmaports/-/merge_requests/4541
170189393b (device-nokia-n900: add upower support and config (MR 4397) added
an upower config, but did not provide a subpackage to actually install it on
device. Add the needed subpackage.
[ci:skip-build]: already built successfully in CI
(1) the deviceprofile differs from upstream 1.16.3 in the following respects:
(a) -export SXMO_NO_KEYBOARD=1 - allow use of the virtual keyboard
(b) +export SXMO_DMENU_LANDSCAPE_LINES="10"
+export SXMO_DMENU_PORTRAIT_LINES="6" - these two adjust the size of the
displayed dmenu menu.
* NB: All foregoing changes are already merged in sxmo-utils master.
(c) +export SXMO_TOUCHSCREEN_ID="TSC2005 touchscreen" - allows sxmo to
reliably toggle touch screen's enabled status
* The foregoing change is awaiting acceptance by upstream.
(2) The start hook differs from upstream 1.16.3 by the removal of the line that
starts conky. That program pegs N900 CPU usage at 100% all the time. This
change is not upstreamable, at least not for the time being
At some point the driver started using symlinks w/ machine IDs for
finding/loading fw, and for some reason I thought it was a good idea to
not install those symlinks.
Fixes this failure when loading the brcmfmac driver:
brcmfmac mmc1:0001:1: Direct firmware load for brcm/brcmfmac43752-sdio.purism,librem5r4.bin failed with error -2
[ci:skip-build]: already built successfully in CI
With the current kernel, ds2482 w1<>i2c module is not automatically
loaded at boot, which prevents loading the ds2784 battery module.
Signed-off-by: Alexandre Marquet <tb@a-marquet.fr>
[ci:skip-build]: already built successfully in CI
This fixes op6 display panel support, drm error spam in dmesg, and
various other minor things.
It also enabled dynamic debugging and ftrace.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
For now, limited support without graphics. Graphics
support would need messing and detecting board variants.
especially since Rev B seems to need a vcom setting per batch.
[ci:skip-build]: already built successfully in CI
Add support for the Xiaomi Miatoll phones. There are currently two
device trees in the upstream kernel.
Signed-off-by: David Wronek <david@mainlining.org>
The original maintainer unfortunately has bricked their phone. Drop the
device package in preparation for the SM7125 kernel upgrade.
Signed-off-by: David Wronek <david@mainlining.org>
The device trees for this phone have never been upstreamed
unfortunately. Drop the device package in preparation for the SM7125
kernel upgrade.
Signed-off-by: David Wronek <david@mainlining.org>
These modules are required in the initramfs when booting a kernel that
builds them as modules (i.e. not builtin)
[ci:skip-build] already built successfully in CI
Since 6.10-rc1 python3 is now a hard requirement to compile the kernel,
it's used to generates headers for drm/msm. Add it now so our CI can
keep working.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
He did an awesome job adding an initial port for this device but hasn't
helped to maintain it, so I'm removing him from the list of active maintainers.
And remove the compatibility symlinks again, all kernels running now
should use the .mbn files already.
[ci:skip-build]: already built successfully in CI
Copied from qemu-amd64. Enabling support for EFI boot allows me/others
to easily test pmOS EFI boot loader and config changes. Legacy BIOS boot
will still work, see 22c602096 for details.
[ci:skip-build]: already built successfully in CI
- 0003-dts-pinetab-add-missing-bma223-ohci1.patch
The original patch added a node for BMA223 and a node for ohci1.
The former is already present in upstream, so the new patch
0003-dts-pinetab-add-missing-ohci1.patch only adds the latter.
- 0004-arm64-dts-allwinner-Add-bluetooth-node-to-the-PineTa.patch
The patch is now present in upstream so it has been removed.
However it's worth noting that upstream's node has the wake GPIOs
reversed from the original patch. The original patch added:
device-wake-gpios = <&r_pio 0 6 GPIO_ACTIVE_LOW>; /* PL6 */
host-wake-gpios = <&r_pio 0 5 GPIO_ACTIVE_HIGH>; /* PL5 */
... while upstream has:
device-wake-gpios = <&r_pio 0 5 GPIO_ACTIVE_LOW>; /* PL5 */
host-wake-gpios = <&r_pio 0 6 GPIO_ACTIVE_HIGH>; /* PL6 */
This has not been tested on a PineTab.
- 0005-dts-pinetab-make-audio-routing-consistent-with-pinep.patch
The first hunk of the patch is in upstream and has been removed.
- 0013-Revert-usb-quirks-Add-USB_QUIRK_RESET-for-Quectel-EG25G-Modem.patch
This patch reverted upstream commit that was present until v6.7
( f9e3e7ea4e )
but is not present in v6.8+, so the patch has been removed.
- 0014-usb-serial-option-add-reset-resume-callback-for-WWAN.patch
The patch is now present in upstream so it has been removed.
- 0016-Define-orientation-and-rotation-for-PinePhone-camera.patch
The patch is now present in upstream so it has been removed.
---
The issues found with the previous upgrade to 6.8.4
( 646e0b50df ), ie reliably receiving calls
in suspend ( #2734 ) and the headset / headphones detection being broken
because of the jack detection changes ( #2738 ) have been fixed.
[ci:skip-build]: already built successfully in CI
Adjust the sensor firmware paths for hexagonrpcd 0.3.1.
Also drop board-2.bin file which is now upstream and use upstream mba.mbn
and wlanmdsp.mbn.
[ci:skip-build]: already built successfully in CI
This causes a lot of spam in dmesg about "failed to allocate compressor transform", making it nearly impossible to search anything in the system logs.
This option has nothing to do with zram.
[ci:skip-build]: already built successfully in CI
Contains fix for Poco F1 display issue when booting and enables sensor
support for Poco F1.
[ci:ignore-count]
[ci:skip-build]: already built successfully in CI
- Enable KVM support (for using with slbounce!)
- Disable RT_GROUP_SCHED, community device kernel requirement
[ci:skip-build]: already built successfully in CI
These kernels must have NLS_ASCII set after bb6d7a05b, or else they
cannot mount the ESP as /boot and boot fails. This kconfig option will
be added to the pmb efi kconfig check so that this is caught in the
future pre-merge.
fixes 3909b1dbf and bb6d7a05b
[ci:skip-build] already built successfully in CI
Unmaintained is a name which on multiple occasions have seen lead to
confusion with people having the impression that unmaintained is for all
unmaintained devices, which is not how we're really using it. Many
devices in testing do not actually have a maintainer, yet there has been
no push to move these out of there and into unmaintained.
I think this is a result of that unmaintained was introduced not to keep
unmaintained ports but rather a place to store ports that have a better
replacement but where the inferior one still holds some sort of value,
such as for debugging purposes. These ports also are not necessarily
entirely unmaintained and see more fixes than many ports in testing.
While one approach to solving this problem could be to simply moving all
unmaintained ports to unmaintained, I think this comes with some
problems: It would require an initial effort to figure out which ports
are indeed unmaintained and which just don't have a maintained noted in
the package, and given how many ports there are in testing this would be
a big endeavour. It would also require continuous work on moving ports
into unmaintained as the maintainers go silent if we are to keep testing
and unmaintained's state consistent with reality. Additionally, just
because a port doesn't have a maintainer on paper doens't mean that
there aren't people who aren't willing to fix it up if there are issues
that arise.
As such, I think the way to go is renaming unmaintained to better
reflect the original intent. Thanks to Luca Weiss for suggesting
"archive", and to Arnav Singh for suggesting that "archived" would match
the other category names better.
