This application doesn't work 100% when launched from the 'desktop', but
it is still useful as a handler for opening image files from a file
browser, so the icon is hidden.
eog is a nice image viewer that works pretty well on a
touchscreen/mobile display.
Nemo is a file manager from the Cinnamon desktop project, which forked
from Nautilus some time ago. It is much more usable than Nautilus on a
touchscreen/mobile display.
Set eog as default app in mimeapps.list, so it even works when hiding
the launcher with postmarketos-hidden-desktop-entries (we'll hide it,
because it doesn't work when opened directly; no image is shown and
opening a file does not work either). The list of associated mimetypes
is taken from here:
https://gitlab.gnome.org/GNOME/eog/-/blob/master/data/org.gnome.eog.desktop.in.in
We now trigger on files in /usr/share/postmarketos-mkinitfs-triggers.
This can be used by other packages like osk-sdl for which an updated
version would otherwise never be used in the initramfs
Enable the freezer cgroup. This allows a user to start a large application like
firefox, freeze it to save some power, and unfreeze it to continue using it,
without having to restart the application completely. Stopping and restarting
the application completely results in more battery usage as well as lost state.
Signed-off-by: Serge Hallyn <serge@hallyn.com>
Sloppy patch that makes the displayport alt-mode output signal work on
the PinePhone convergence edition. Proper fix is in the 5.8 tree
[ci:skip-build]: Martijn made sure that it builds and works. This is
time critical, so skip the CI build.
Fix mobile-config-firefox not replacing postmarketos-config-firefox on
upgrade. Without the pkgver in provides, apk only considers
mobile-config-firefox as virtual package name, not as replacement.
Verified with the CE factory image:
$ sudo apk upgrade
...
(20/29) Purging postmarketos-config-firefox (1-r0)
(21/29) Installing mobile-config-firefox (1.0.1-r1)
...
postmarketos-ui-plasma-mobile has plasma-camera in _pmb_recommends.
plasma-camera does not build for x86, because mpg123-libs-1.26.3-r0 is
still broken in Alpine edge repos (aports#11807).
Disable postmarketos-ui-plasma-mobile for x86 again, so the CI check
test_aports_ui doesn't fail anymore with:
postmarketos-ui-plasma-mobile/APKBUILD: package 'plasma-camera' from _pmb_recommends not found for arch 'x86'
[ci:skip-vercheck]: only modifying arches
[ci:skip-build]: unfortunately, postmarketos-ui-plasma-mobile fails to
build for x86_64 too due to broken libdv in Alpine (aports#11892).
I'll create MRs to Alpine to fix the broken depends, so we should be
able to enable it again soon.
Related: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11807
Related: https://gitlab.alpinelinux.org/alpine/aports/-/issues/11892
Pressing power button should not power off a device. Although it is
expected on desktops and laptops, it isn't expected on phones and
tablets. Instead it should just turn off the screen or go into a deep
sleep, but that isn't handled by elogind. So let elogind just ignore the
button altogether.
Also re-order the dependencies alphabetically while we're touching the
package anyway
Previously we added the header in the kernel packages but as we already
add the ramdisk header here and to simplify the process we also add the
addition of the kernel header into mkinitfs.
With the new mobile-config-firefox, the window width is small enough
that the window does not need to be scaled. This looks better overall,
and avoids scaling glitches.
New version does not just support Firefox 68 (firefox-esr in Alpine 3.12),
which is going to be EOL soon. It also supports the new ESR 78
(firefox-esr in Alpine edge, soon also in 3.12, supported until ~2021-04)
and 79 (current stable "rapid release" version; firefox in Alpine edge).
The source was moved to a dedicated git repo, so other distributions can
also use it. Keep the distro-specific links here in a separate HTML
file, so they can be modified without changing
mobile-config-firefox.git.
Related: https://gitlab.com/postmarketOS/mobile-config-firefox
- Update to Linux 5.8
- samsung-skomer:
- Fix accelerometer mount-matrix
- Fix some potential display issue(s)
- Enable touchscreen driver in kernel config
[ci:skip-build]: already built successfully in CI
See the patch descriptions for more details. All patches will be
submitted upstream (patch 0001 has already been submitted more than half
a year ago but hasn't had any activity yet).
"plasma" and some related packages are currently missing in Alpine edge
armv7 due to a cyclic dependency. Let's disable all packages depending
on plasma for armv7 temporarily, to get the pmOS edge armv7 repo up
again.
[ci:skip-build], [ci:skip-vercheck]: only arch line changed
Related: build.postmarketos.org#72
Add mimeapps.list, so links clicked in postmarketos-welcome and other
applications properly open in firefox. Without this patch, they just
don't open at all.
Bump the pkgver instead of the pkgrel, so the APKBUILD is in sync again
on both master and v20.05 (right now, the pkgrel on v20.05 is ahead by
one, otherwise the file is the same).
Related: https://help.gnome.org/admin/system-admin-guide/stable/mime-types-application.html.en
Disable the first run screen in GNOME software, which invites the user
to browse and install apps. We do not want people to use GNOME software to
do that when they open it for the first time in postmarketOS after
installing, we want them to do updates. So this dialog is misleading for
our users, and will lead to a bad experience for those who follow what
it suggests.
I think there's more wrong with that dialog (in context of postmarketOS),
more detailed reasoning and screenshot in:
https://gitlab.com/postmarketOS/pmaports/-/merge_requests/1449#note_385456592
Replace firefox with firefox-esr, remove epiphany. With the current
version of epiphany, the postmarketOS wiki doesn't even load every time,
so that's not something we want to ship.
Add policies.json and prefs.js to configure Firefox (yes, you need both
to be able to control all this):
Mobile improvements:
* Enable zooming with fingers
* Mobile user agent (from tor browser for android)
Privacy improvements:
* Disable search suggestions, so URLs do not get sent to search engines
as they are getting typed.
* Disable Firefox studies
* Disable Telemetry
* Set DuckDuckGo as default search engine, not Google
Uncluttering (screen space, remove broken features, less distractions):
* Move all buttons to the overflow menu and remove spacers around the
address bar
* Empty "new tab" page (loads faster, no annoying "top sites" etc.)
* Disable developer tools, so the hamburger menu fits the screen of the
pinephone. These aren't really useful on phones anyway.
* Disable "Firefox Screenshots": the feature did not work with the mobile
resolution in Phosh.
* No default bookmarks from Firefox
* Disable First Run Page
* Disable Post Update Page
* Disable "User Messaging" (What's new, Extension/Feature
Recommendations, Urlbar Interventions)
Add a local html file that serves as lightweight default homepage. It
links to the postmarketOS blog and wiki, explains that this is desktop
firefox with mobile configuration and explains how to change settings
and install addons (with a direct link to ublock origin). Users can
override the homepage, new tab page, search engine, search suggestions
pre-configured by this package directly in the UI.
I chose Firefox ESR, because:
* It fits the screen better: the "new tab" button is bigger, the menu
opening when you click the top-right button actually fits the screen,
with all its submenus. There are zoom controls in the same menu, which
make the preferences pages very usable.
* It's possible to override the default search engine with a policy
file; that's not possible in non-ESR firefox (we'd probably need to
change it in the sources).
Related: https://github.com/mozilla/policy-templates
Co-authored-by: Martijn Braam <martijn@brixit.nl>
In phosh, we have broken Cheese (Camera) and Extensions (from GNOME shell)
icons in the launcher. They get dragged in via dependencies, so the
easiest way to disable them is just overruling the .desktop icon. We can
make Cheese visible again in a post-install script (just check if it
points to nodisplay.desktop and delete the link) once we have it working.
Hide icons for terminal programs (htop, nvim, vim) as well as "Firefox
Safe Mode" too. If somebody really wants to use the FF safe mode, they
can do so via terminal.
Have a simple initial version, maybe make it more sophisticated with
UI-specific hidden icons later (right now, it doesn't make sense to
install this in normal GNOME, because there you would need
org.gnome.Extensions). (There is "NotShowIn=phosh" (untested), but using
this would mean that we need to keep all the information of the original
desktop file for the UIs that will display them; we can't just use the
symlink trick. I'm not sure if we want that at all, but it's
definitively over-engineered for now.)
The p? is not redundant. Without it, the p in mmcblk0p2 for example does
not get cut off, meaning the resulting device is not "mmcblk0" but
"mmcblk0p". My bad, sorry for the breakage.
Fixes: 9f6600ba ("main/postmarketos-mkinitfs: rm pmOS_deleteme")
Fixes: 9d86f6fe ("main/postmarketos-mkinitfs: resize: unallocated space check, even if forced")
Remove the "pmOS_deleteme" partition left behind by the on-device
installer, if it exists. Let the existing resize_root_partition() extend
the root partition over the newly gained space (and rest of the storage
device) right afterwards.
In resize_root_partition(), put the unallocated space check into an
extra function has_unallocated_space(). Run it even if
PMOS_FORCE_PARTITION_RESIZE is used, so we don't attempt to resize the
partition if there is no free space.
While at it, change "sed 's/p?2$//'" to "sed 's/2$//' in an already
modified line, because it means the same thing and is less confusing
(? in regex means: 0 or 1 time).
Enable CONFIG_CRYPTO_XTS for each kernel, so we can switch to using
aes-xts-plain64 as default cipher for cryptsetup (override with
"pmbootstrap --cipher"), instead of aes-cbc-plain64 (pmbootstrap#1940).
I have executed "pmbootstrap kconfig edit" on each kernel, and manually
toggled the option. The diff is not always clean, because for some
kernels it is apparently the first time, that menuconfig was executed on
the configs like that. In a few instances, it turned out that
CONFIG_ANDROID_PARANOID_NETWORK needed to be disabled too (this is
already a requirement, but as the config was incomplete, it was not
visible that this option was enabled). Very few times, I had to enable
CONFIG_EXPERIMENTAL in order to see and enable CONFIG_CRYPTO_XTS.
It would be great if we could automate such mass kconfig edits in the
future, see pmbootstrap#1942.
[skip ci]: I have verified, that every single one of these kernels builds.
CI will likely run out of time while downloading source tarballs.
Seems like "pmbootstrap kconfig edit" causes unrelated kernel config
changes for some reason (probably because pmbootstrap does not install
a cross compiler for it). For now let's just edit the config manually
with the new options so everything else stays as-is.
Only add a comment, do not try to mount the root or boot partitions
anymore. The initramfs is doing this already.
Related: https://postmarketos.org/fstab
Mount the boot partition at /sysroot/boot and keep it mounted, when
running the switch_root command. This way, OpenRC doesn't need to mount
it and possibly use the wrong partition. The OpenRC service does not use
the same logic to find the boot partition, in particular it does not
support the pmos_boot kernel parameter.
While at it, print the mountpoint and read-only/read-write arguments in
the mounting log message for both root and boot.
Fixes: #664
Get rid of the "write_unless_modified" code, which was supposed to only
change /etc/fstab, /etc/issue, /etc/motd if the user did not modify it.
This is nice in theory, but we have a bug report where the code did not
do what it should (apk audit possibly failed due to a qemu bug?), and
then it lead to strange bugs related to not having the expected
/etc/fstab installed.
Fixes: #661, #258
This enables mult-sensor support in sun6i-csi, support for the
gc2145 front camera, the new panel driver from mainline linux
and a small fix for the anx7688 that prevents it spamming the
kernel log when no cable is plugged in on 1.2a hardware.
gitlab.com had a service disruption, so it was not possible to "git
clone" from gitlab.com from some regions, like where the sourcehut
infrastructure is located. Since we are building our packages on
builds.sr.ht, this caused all packages that should be built at that time
to fail. It is working again, so bump the pkgrels to restart the builds.
This switch has happened in the Alpine repos quite a while ago and most
of the pmOS packages were using it already too, so let's switch over the
last ones as well.
This also cleans up the APKBUILDs where necessary
In the Linux kernel, the ST-Ericsson NovaThor U8500 SoC is barely
ever mentioned under the "NovaThor" name, more common are names like
U8500, Ux500, ST-Ericsson, STE, ...
Let's rename the kernel package to linux-postmarketos-stericsson
to make that more clear. Also cleanup the APKBUILD a bit while we're
at it, HOSTCC is unneeded, remove unnecessary depends, ...
We cannot have two of them at the same time, most compiler would fail here
but for some reason our didn't.
However, it's better to just use one (the new one, or the old one). Enable
CONFIG_RTL8723CS_NEW, which is the new driver that made it to pine64
kernel since 5.5.
Signed-off-by: Danct12 <danct12@disroot.org>
[ci:skip-build]: already built successfully in CI
We should disable it by default and let the user to enable it
when they want it.
And currently it's causing problems where the backlight would
go very dark with it enabled on PinePhone.
Signed-off-by: Danct12 <danct12@disroot.org>
This release gets rid of the hacks needed for out-of-tree modules in the
previous version of calamares.
[ci:skip-build]: already built successfully in CI
This makes recording system audio very simple and feasible in a cross-device
manner since application-logic can always rely / forward ALSA audio to the
loopback device instead of a device-specific audio interface.
We are in the midst of the Linux 5.8 merge window so there is no new
-rc version to rebase on. -rc1 next week is also not the perfect choice
for a release (first rc is usually still a bit unstable).
On the other hand the changes that landed the last few days are quite
nice and ready to go, so I decided to rebase on a stable kernel release
(5.7.1). Not sure if I'm ever going to do this again :)
Changes:
- Update to Linux 5.7.1
- bq-paella: Add sound and modem
- Enable some new kernel options for DB410c
dbus-launch doesn't get installed automatically. On the other hand,
dbus-run-session does. This version ensures that dbus is also launched
for the graphical demos
Disable modules for hardware that's impossible to connect to the
allwinner devices or older than the 90's this shrinks the size from 380
to 250MB.
Then it enables module compression so it shrinks from 250MB to 57MB.
Then the module stripping was enabled, shrinking it to 20MB
[ci:skip-build]: already built successfully in CI
* Support for sharp ls052t3sx02 video mode panel and max1187x
touchscreen used in Sony Xperia Z2 (sirius)
* Support for display in OnePlus One (bacon)
* Support for touchscreen, touchkeys and notification led in Samsung
Galaxy S5 (klte)
At the moment we have Contributor: lines on some packages (but not all of them),
but often they don't represent the actual contributors to the package very well.
E.g. when we added them retroactively to the device packages we only added
the initial contributor (which isn't necessarily the person
who made most of the work for a device...)
The Git history is the most representative source for figuring out
who contributed to a package, so there is no reason to duplicate that
into the APKBUILD.
[skip ci]: way too many packages
fixes the desktop file:
* add -- to make cage ignore kodi's arguments
* add --standalone to kodi's arguments
update the dependencies:
* add upower as kodi tries to communicate with it to read battery status
* add pulseaudio as kodi complains it is missing
* add elogind as it is required for cage to run
* remove xorg-server
Fixes#469
- Update to Linux 5.7-rc2
- Initial support for BQ Aquaris X5 (bq-paella/picmt)
- Add speaker for samsung-gt510wifi/matissevewifi
- Add flash LED for wileyfox-crackling
Read kernel module names from files in
/etc/postmarketos-mkinitfs/modules and print out which modules are
required by which of these files as the initramfs gets generated. Put
the default modules into a new 00-default.modules. This allows mkinitfs
hooks to add modules to the initramfs.
Parse the modules from a file similar to modules-load.d, with commented
lines and empty lines ignored. Add a simple test script for that and run
it in check().
Split mkinitfs.sh into two files, so the functions (now in
mkinitfs_functions.sh) can be sourced in a test. Such a test will be
added in the next commit.
Move $BINARIES_EXTRA into the get_binaries_extra() function, so its
$(find...) does not get executed when sourcing the functions file. Move
$BINARIES into get_binaries() too for consistency.
Make it pass shellcheck by fixing obvious mistakes like '==' instead of
'='. Add "shellcheck disable" lines for things that are intentionally done
or would be a bit more complicated to rewrite. It would be nice to
refactor some things like '${outfile/initramfs-/uImage-}', but this should
be done in a separate change (patches welcome).
Remove unused variable "modules_path". While at it, add two new fatal
error messages instead of just "exit 1" without any comment and remove
redundant mkdir in APKBUILD.
We'll use the new RTL8723CS driver in the kernel tree instead
of using the one that we have to fetch from GitHub.
Signed-off-by: Danct12 <danct12@disroot.org>
Signed-off-by: Asriel Dreemurr <asriel.danctnix@gmail.com>
[ci:skip-build]: already built successfully in CI
The lddtree script uses the scanelf utility which has some security check that causes a slowdown of the operation, executing it without privileges it's really faster.
The current postmarketOS port does not do Hildon/Maemo Leste justice.
Various features are missing or broken. We can't even upgrade to the newest
version, because of packaging issues that can't easily be resolved. And
nobody is maintaining it currently. So for the people who want to use
Hildon, please use Maemo Leste (https://maemo-leste.github.io/).
If somebody is willing to invest a lot of time for a new postmarketOS
port of Hildon/Maemo Leste, let us know!
[skip-ci]: just downloading all kernel sources takes about one hour
and ollieparanoid wants to merge it now. This already ran
successfully with [ci:skip-build] and [ci:skip-vercheck].
Many device ports install extra modules-load.conf/modprobe.conf
configurations. Simplify this a bit by installing these automatically
to appropriate locations in the file system.
Left over from unity 8, which has been removed from pmaports earlier.
umockdev depends on libgudev in Alpine, which has been moved from main
to community. As of writing, there is no armhf package, so the upstream
compatibility test is failing. There is no point in waiting until the
package was built though, since nobody is currently using in in
pmaports.
CONFIG_INPUT_UINPUT is needed for !1012
- linter fix: remove unneeded HOSTCC
- use downstreamkernel_package
- _abi_release was never used
- this package is only enabled for aarch64, so always use
modules_install dbts_install in package(), remove redundant
switch case
The point of this program was to have something to play with once weston
boots up. It does not make sense to ship this with
postmarketos-ui-plasma-mobile.
Related: postmarketos-demos#1
nokia-n900 is the last user of linux-postmarketos-stable.
Let's keep it as-is temporarily until nokia-n900 has been migrated
to Alpine's linux-lts or a custom kernel.
Disable it on all other architectures to make it clear
that it is obsolete.
[ci:skip-vercheck]: no changes, only disabled architectures
- Update to Linux 5.6-rc5
- Basic audio support on samsung-gt510wifi (speaker is WIP)
- Initial support for samsung-matissevewifi
- Enable some kernel config options for elogind/Anbox/fbkeyboard
These dependencies don't belong here.
- iw: should be installed by the user themselves when wanted
- nano: should be installed by the user themselves when wanted
- pulseaudio: should be installed by the UI meta packages that need it
instead
- pulseaudio-alsa: should automatically be installed when both
PulseAudio and alsa are installed
https://gitlab.alpinelinux.org/alpine/aports/merge_requests/4820
Partial revert of 24fa68c9 ("avoid storing initramfs-extra twice").
Some people have (rightfully) complained that including the hash
in the filename of the initramfs-extra prevents the device from booting
whenever the boot partition gets out of sync with the boot image.
In general we should assume that those two are in sync, but there is
no need to unnecessarily break this if previously booting the outdated
extra initramfs worked just fine.
At the end, the hash is just an implementation detail for the caching
mechanism - we need it when (re-)generating the initramfs, but it does
not have to be within the filename. Instead, we can simply append the
hash to the file contents. GZIP will simply ignore the trailing garbage.
The boot partition will then look like:
$ cat /boot/initramfs-<flavor>-extra
<GZIP data>d3c7b449c6fc811d97351bbc46852b66 (the hash)
This makes the filename of initramfs-extra nice and stable again.
It also fixes a regression where the initramfs-extra would no longer
get symlinked by "pmbootstrap export" (I was going to fix that in
pmbootstrap but now this solution is more appealing...)
While we're at it: Make the script more resilient again by writing
to a temporary file first, then move it (atomically) to the real path.
We should do this in the UI packages that need it rather than in the
device packages themselves because for example Phosh doesn't need Ofono
but ModemManager
HOSTCC defaults to gcc, so we should only set it to override it
if we want to use an older GCC. Removing it avoids annoying linter
warnings caused by the HOSTCC="..." variable assignments.
Currently, every LCDs inside their (our users) pocket are driving 3.3V,
meanwhile the LCD interface is meant to drive at ~2V.
This turns our LCDs into a timebomb that is ready to trigger at any moment,
killing the display.
Signed-off-by: Danct12 <danct12@disroot.org>
- Update to Linux 5.6-rc2
- Major improvements for Motorola Moto G4 Play (motorola-harpia)
- Display
- Touchscreen
- Accelerometer
- Audio
- Modem
- Vibrator
- Fix headphones/headset detection for wileyfox-crackling
- Add mount-matrix for samsung-a3ulte/a5ulte to fix
orientation detection with iio-sensor-proxy
- A lot of cleanup everywhere
- APKBUILD: Clean up makedepends (a lot of them are unneeded)
wileyfox-crackling was the only user of QCDT. Now that it is also
using appended DTBs, there is no need to generate the QCDT image anymore.
Another advantage of this is that mainline can then be only booted using lk2nd
(as intended), since booting it from the stock bootloader has weird side effects
(e.g. no WiFi/BT MAC address set).
That path works just as well as documented at
https://doc.qt.io/qt-5/qstandardpaths.html (GenericConfigLocation) and
it doesn't depend on the user being created after the package has been
installed.
polkit rely on systemd, so we don't get correctly working setup.
This impact non-working backlight control from userspace and other
stuff.
Signed-off-by: David Heidelberg <david@ixit.cz>
At the moment we store the same initramfs-extra twice on the file system:
- /boot/initramfs-<flavor>-extra
- /var/cache/postmarketos-mkinitfs/initramfs-<flavor>-extra_<hash>
The second copy is exactly the same file, just with the <hash> (for caching)
appended to the filename.
We can avoid this by appending the hash directly to the filename
on the boot partition, i.e.
- /boot/initramfs-<flavor>-extra_<hash>
This is possible because we only reference initramfs-extra from the main
initramfs, and we already replace the path in it dynamically.
It will just let the main initramfs load -extra_<hash> instead of just -extra.
This saves a few megabytes of disk space on the rootfs.
postmarketos-ui-sway already depends on xorg-server-xwayland for
X clients. There is no need to pull in the entire X server including
all necessary drivers.
This saves ~84 MiB of disk space when installing postmarketos-ui-sway
on asus-me176c. X applications are still working fine through Xwayland.
Also remove explicit dependency on dbus - it is already pulled in
by dependencies like lightdm or elogind and nothing in
postmarketos-ui-sway depends on it specifically.
We need to generate the splash screens separately for each device,
because they are specific to the device's display resolution.
At the moment we do this dynamically during the installation process.
This has the advantage that there is no need to re-build all device
packages when one of the splash screen is changed (or a new one is added).
In reality, however, the splash screens do not change very frequently.
On the other hand, generating the splash screens dynamically has signficant
disk usage overhead for a minimal ("none" UI) rootfs:
The Python interpreter together with the necessary libraries requires
about ~60 MB of disk space on aarch64.
The splash screens itself require about ~100 KB for 720x1280.
This is not necessary if we move the splash screen generation into
devicepkg-dev, which is used to build the device package for all devices.
Another advantage is that we no longer need the (rather complicated)
caching mechanism for splash screens - so we actually end up with less
lines than before.
rootfs size for samsung-a5ulte ("none" UI):
Before: 450M
After: 388M (-62M)
After this change, every(!) device package needs to be rebuilt once.
No changes are necessary in device packages.
Kernel subpackages may contain a "-" (e.g. -kernel-mainline-modem);
replace it with an underscore to make sure that we find the
kernel-specific variables in deviceinfo.
If (a) the kernel command line parameters include PMOS_BOOTCHART2 and
(b) /sbin/bootchartd exists (i.e., the bootchart2 package has been
installed), run /sbin/bootchartd instead of /sbin/init.
Port bootchart2 to PostmarketOS, with three patches:
1. Modification of /etc/bootchartd.conf so that the collector
will stop after sshd has started.
2. Python 3.8 compatibility
3. Disable the interactive mode of pybootchartgui.
Use bootchart2 to profile system boot or a running system.
1. Add the bootchart2 package to your device.
2. Add the following kernel command line parameters: initcall_debug
printk.time=y quiet
3. To profile system boot
a. add the following kernel command line parameter: PMOS_BOOTCHART2.
This parameter causes mkinitfs to start /sbin/bootchartd instead of
/sbin/init. /sbin/bootchartd then starts /sbin/init.
b. (optional) modify EXIT_PROC in /etc/bootchartd.conf to define
when the collector is to stop.
c. Boot the system. Approximately 20 seconds after sshd starts
(if using the default value of EXIT_PROC), the collector
will exit.
4. To profile a running system
a. use the command '/sbin/bootchartd start' (as root) to start
profiling
b. use the command '/sbin/bootchartd stop' (as root) to stop
profiling
5. When the collector exits, /usr/bin/pybootchartgui will create
a visualization in /var/log/bootchart.png.
- Update to Linux 5.5-rc4
- Fix freeze/crash when accessing reserved memory on Samsung devices
- This happened e.g. when doing large eMMC writes
because the disk cache filled all available RAM.
- Fix SD card detection on samsung-serranovelte
- Disable UHS-1 modes since they appear to be broken in hardware
[ci:skip-build]: already built successfully in CI
Comment out the initstepslew line in /etc/chrony/chrony.conf.
initstepslew can delay booting while chrony sets the time. The
makestep 1 -1 line in the configuration also causes chrony to step
the time (whenever the offset is greater than 1s), making
initstepslew redundant.
Unfortunately, early firmware versions on MSM8916 (used together with
Android 4.4) are not capable of booting aarch64 kernels. MSM8916 was
Qualcomm's first 64-bit capable CPU, and back when it was introduced
Android did not even support aarch64 yet. So the aarch64 capable
hardware was usually used together with armv7 binaries.
To make things worse, this firmware is also signed (usually per-device),
making it (currently) impossible to update it without help from the manufacturer.
This means that we are currently unable to run mainline on devices
that never received an update to Android 5.0
(e.g. the Samsung Galaxy S4 Mini Value Edition).
One way to circumvent this limitation is to build mainline for armv7.
This works surprisingly well with some additional fixes.
Devices with this firmware limitation are still unable to use aarch64
binaries, but at least we can run mainline on them!
- Rebase on Linux 5.5-rc3
- Add display, touchscreen and GPIO keys for samsung-gt510wifi
- Add vibrator for wileyfox-crackling
- Zinitix touchscreen improvements (used in samsung-a3ulte)
- Upstreamed several patches
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
