OmniROM Android R (11) for Pi 4

Search This thread

rabbited

Member
Mar 10, 2020
37
5
I am not a developer and am using Raspberry Pi as a hobby device. This was meant to be a bug report as I live in NYC which has a population of 9 million and I can't use OmniRom WiFi anywhere in the city so far due to public Hotspot 2.0, so I assume that no one here can. I assume this will be a problem in other cities and since there is very little discussion on the topic and none of the developers of different flavors of Android for RPi seem to experience this in their locale I would say that most people either blame the bootloop on something else or have no idea what causes the problem and give up on any Android on RPi. Here is a thread in another group showing a few logcats where the device reboots immediately following a Passpoint query:


This is not an OmniRom specific issue but the developer of Android-rpi is not responding to questions on his github.

I'll keep working on trying to figure out how to edit device.mk or another file to disable Passpoint to share with the community, I just figured you may have an interest in shutting off a potentially broken feature. One common thread is that every user seems to have an Altice WiFi query right before reboot, so perhaps it's not Passpoint overall but something about Altice's network that causes the issue? If that's the case it only effects users in NY, NJ, CT and the other two states this company serves.

If you'd still like a logcat I'm glad to generate my own. I'll keep trying to sort out how to edit the proper files and share if I figure anything out.
 

rabbited

Member
Mar 10, 2020
37
5
I added two logcats. One I did over adb which shows the messages I'm referring to. The other is through Matlog and doesn't contain the same info so I'm not sure it retained the pre-reboot log but can try again tomorrow if Matlog is necessary.

 
Last edited:

maxwen

Senior Member
Jun 10, 2012
8,068
10,285
I am not a developer and am using Raspberry Pi as a hobby device. This was meant to be a bug report as I live in NYC which has a population of 9 million and I can't use OmniRom WiFi anywhere in the city so far due to public Hotspot 2.0, so I assume that no one here can. I assume this will be a problem in other cities and since there is very little discussion on the topic and none of the developers of different flavors of Android for RPi seem to experience this in their locale I would say that most people either blame the bootloop on something else or have no idea what causes the problem and give up on any Android on RPi. Here is a thread in another group showing a few logcats where the device reboots immediately following a Passpoint query:


This is not an OmniRom specific issue but the developer of Android-rpi is not responding to questions on his github.

I'll keep working on trying to figure out how to edit device.mk or another file to disable Passpoint to share with the community, I just figured you may have an interest in shutting off a potentially broken feature. One common thread is that every user seems to have an Altice WiFi query right before reboot, so perhaps it's not Passpoint overall but something about Altice's network that causes the issue? If that's the case it only effects users in NY, NJ, CT and the other two states this company serves.

If you'd still like a logcat I'm glad to generate my own. I'll keep trying to sort out how to edit the proper files and share if I figure anything out.tl;

I am not a developer and am using Raspberry Pi as a hobby device. This was meant to be a bug report as I live in NYC which has a population of 9 million and I can't use OmniRom WiFi anywhere in the city so far due to public Hotspot 2.0, so I assume that no one here can. I assume this will be a problem in other cities and since there is very little discussion on the topic and none of the developers of different flavors of Android for RPi seem to experience this in their locale I would say that most people either blame the bootloop on something else or have no idea what causes the problem and give up on any Android on RPi. Here is a thread in another group showing a few logcats where the device reboots immediately following a Passpoint query:


This is not an OmniRom specific issue but the developer of Android-rpi is not responding to questions on his github.

I'll keep working on trying to figure out how to edit device.mk or another file to disable Passpoint to share with the community, I just figured you may have an interest in shutting off a potentially broken feature. One common thread is that every user seems to have an Altice WiFi query right before reboot, so perhaps it's not Passpoint overall but something about Altice's network that causes the issue? If that's the case it only effects users in NY, NJ, CT and the other two states this company serves.

If you'd still like a logcat I'm glad to generate my own. I'll keep trying to sort out how to edit the proper files and share if I figure anything out.
Don't write essays - get me a logcat
Ofc created with the provided matlog app
 

jigblaz

Member
Feb 23, 2018
19
3
omni 12 twrp not opening after setting boot mode to recovery, now i cant do anything
i even tried formatting and reflashing multiple times but same thing happens every time
 

maxwen

Senior Member
Jun 10, 2012
8,068
10,285
omni 12 twrp not opening after setting boot mode to recovery, now i cant do anything
i even tried formatting and reflashing multiple times but same thing happens every time

well - after stating the obvious - it works for me
SD or USB boot?
Do you still get adb connection?

BTW: just mount on PC and copy the config.txt.rom to config.txt then it will
be again boot the ROM
 
Last edited:

maxwen

Senior Member
Jun 10, 2012
8,068
10,285
I added two logcats. One I did over adb which shows the messages I'm referring to. The other is through Matlog and doesn't contain the same info so I'm not sure it retained the pre-reboot log but can try again tomorrow if Matlog is necessary.


Dont change logs - use Matlog and save as zip - I want that and nothing else please
 

rabbited

Member
Mar 10, 2020
37
5
No logs were edited, MatLog seemed not to be saving the pre-crash info in the log. I'm using an edited service-wifi.jar file that removes the ANQP query code and Wi-Fi now works fine.

I'm now having issues with "Start on Boot" in OmniRemote. I am unable to connect via VNC until I toggle the connect button. The status shows as Running but my laptop reports that the connection was refused until i toggle the switch off and back on.
 

maxwen

Senior Member
Jun 10, 2012
8,068
10,285
No logs were edited, MatLog seemed not to be saving the pre-crash info in the log. I'm using an edited service-wifi.jar file that removes the ANQP query code and Wi-Fi now works fine.

Well - you clearly are unable to follow instructions I gave how to create the log so what do you expect me to think of it

I'm now having issues with "Start on Boot" in OmniRemote. I am unable to connect via VNC until I toggle the connect button. The status shows as Running but my laptop reports that the connection was refused until i toggle the switch off and back on.

Most prolly your network is not up fast enough after boot

Another Matlog zip - made correctly - would help
Wait about 30s after boot before creating
 

Quantalou

Member
Jan 7, 2022
6
0
Hello maxwen.
Thank for your great job.
We are using Android11/RPi4 as internal cheap tool for demonstrator and development purposes (not included in a product)
Some modifications have been done to meet our needs:
-configuration as automotive unit (In Vehicle Infotainment) with a new device derived from your RPi4 support
-management of our specific hardware through a HAL and an App dedicated

To be up to date, I started migration to Android12.
I tried to compile our new version based on Android12. I follow similar instruction from Android11.

But, in resulting build, I don't find kernel image and related stuff.

Do I miss something ?
Android12 version is complete and usable ?
 

maxwen

Senior Member
Jun 10, 2012
8,068
10,285
Hello maxwen.
Thank for your great job.
We are using Android11/RPi4 as internal cheap tool for demonstrator and development purposes (not included in a product)
Some modifications have been done to meet our needs:
-configuration as automotive unit (In Vehicle Infotainment) with a new device derived from your RPi4 support
-management of our specific hardware through a HAL and an App dedicated

To be up to date, I started migration to Android12.
I tried to compile our new version based on Android12. I follow similar instruction from Android11.

But, in resulting build, I don't find kernel image and related stuff.

Do I miss something ?
Android12 version is complete and usable ?

Build command is

Code:
m; m kernel

Kernel is not built as part of the main build command

12 is as usable as 11 ever was :)
 

Quantalou

Member
Jan 7, 2022
6
0
Build command is

Code:
m; m kernel

Kernel is not built as part of the main build command

12 is as usable as 11 ever was :)
Hello maxwen

Compilation successful, flashing sdcard ok, run OK: thank you very much
It seems that Android12 is faster than 11. (on RPi4B 8 GiB).

(y)

:D

Question: in device/brcm/rpi4/BoardConfig.mk
TARGET_ARCH := arm64
TARGET_ARCH_VARIANT := armv8-a
TARGET_CPU_ABI := arm64-v8a
TARGET_CPU_VARIANT := cortex-a53

Why do you use cortex-a53 ? BCM2711 is a cortex-a72.

Thank again for you great job.
 

maxwen

Senior Member
Jun 10, 2012
8,068
10,285
Hello maxwen

Compilation successful, flashing sdcard ok, run OK: thank you very much
It seems that Android12 is faster than 11. (on RPi4B 8 GiB).

(y)

:D

Question: in device/brcm/rpi4/BoardConfig.mk
TARGET_ARCH := arm64
TARGET_ARCH_VARIANT := armv8-a
TARGET_CPU_ABI := arm64-v8a
TARGET_CPU_VARIANT := cortex-a53

Why do you use cortex-a53 ? BCM2711 is a cortex-a72.

Thank again for you great job.

The arch values are not really important :)
Important is only arm64 and armv8-a

And it always feels faster if you start fresh :)
But the newer mesa version could indeed make a difference - did not really test
 

Quantalou

Member
Jan 7, 2022
6
0
Hello
On my RPi4/Android11, I am connecting a custom display on HDMI#0. [email protected]
This display is not able to provide EDID so I put in /boot/config.txt.rom (and config.txt) statements to setup this display

[HDMI:0]
hdmi_group=2
hdmi_mode=87
hdmi_timings=1920 0 40 40 48 720 0 4 2 3 0 0 0 60 0 89579520 8
hdmi_force_hotplug=1
[HDMI:1]
hdmi_force_edid_audio=1
[all]

# disable kms audio path and force bcm2835 card 0 = hdmi card 1 = headphone jack
dtoverlay=vc4-kms-v3d-pi4,noaudio
disable_overscan=1
overscan_left=0
overscan_right=0
overscan_top=0
overscan_bottom=0

I tested this config with RaspiOS and it is working well.

On RPi4/Android11, kernel command line contains
video=HDMI-A-1:[email protected]

Framebuffer seems to be right set:
dmesg |grep -i framebuf
[ 0.323709] simple-framebuffer 3e6b4000.framebuffer: framebuffer at 0x3e6b4000, 0x546000 bytes, mapped to 0x(____ptrval____)
[ 0.323729] simple-framebuffer 3e6b4000.framebuffer: format=a8r8g8b8, mode=1920x720x32, linelength=7680
[ 0.323953] simple-framebuffer 3e6b4000.framebuffer: fb0: simplefb registered!

rpi4:/ # ls /sys/devices/platform/3e6b4000.framebuffer/of_node/
compatible format height name reg status stride width
conti_rpi4:/ # od -t x1 /sys/devices/platform/3e6b4000.framebuffer/of_node/height
0000000 00 00 02 d0
0000004
rpi4:/ # od -t x1 /sys/devices/platform/3e6b4000.framebuffer/of_node/width
0000000 00 00 07 80
0000004
rpi4:/ # echo $(( 0x780)) x $(( 0x2d0 ))
1920 x 720

logcat | grep -i surface
01-01 01:04:33.900 4786 4786 I SurfaceFlinger: Using HWComposer service: 'default'
01-01 01:04:33.901 4786 4786 I SurfaceFlinger: SurfaceFlinger is starting
01-01 01:04:33.910 4786 4786 I SurfaceFlinger: Disabling blur effects, they are not supported.
01-01 01:04:33.910 4786 4786 I SurfaceFlinger: Treble testing override: 'false'
01-01 01:04:33.912 4786 4786 W SurfaceFlinger: Unable to open libdolphin.so: dlopen failed: library "libdolphin.so" not found.
01-01 01:04:33.913 4786 4786 E SurfaceFlinger: Unable to open libframeextension.so: dlopen failed: library "libframeextension.so" not found.
01-01 01:04:33.942 4786 4786 I SurfaceFlinger: SurfaceFlinger's main thread ready to run. Initializing graphics H/W...
01-01 01:04:34.081 4786 4786 F SurfaceFlinger: Missing internal display after registering composer callback.
01-01 01:04:34.081 4786 4786 F libc : Fatal signal 6 (SIGABRT), code -1 (SI_QUEUE) in tid 4786 (surfaceflinger), pid 4786 (surfaceflinger)
01-01 01:04:34.164 4812 4812 F DEBUG : pid: 4786, tid: 4786, name: surfaceflinger >>> /system/bin/surfaceflinger <<<
01-01 01:04:34.191 4812 4812 F DEBUG : #03 pc 0000000000110ed8 /system/lib64/libsurfaceflinger.so (android::SurfaceFlinger::init()+3344) (BuildId: 2728199d2c528c4f4dbae73974da8516
)
01-01 01:04:34.191 4812 4812 F DEBUG : #04 pc 00000000000021a8 /system/bin/surfaceflinger (main+352) (BuildId: bb08dae93d29f9e39f6d8131b048a31e)

This is repeating every ~5 seconds

Any idea ?
Somethings to set into display config ?
 

Quantalou

Member
Jan 7, 2022
6
0
Hello
On my RPi4/Android11, I am connecting a custom display on HDMI#0. [email protected]
This display is not able to provide EDID so I put in /boot/config.txt.rom (and config.txt) statements to setup this display

[HDMI:0]
hdmi_group=2
hdmi_mode=87
hdmi_timings=1920 0 40 40 48 720 0 4 2 3 0 0 0 60 0 89579520 8
hdmi_force_hotplug=1
[HDMI:1]
hdmi_force_edid_audio=1
[all]

# disable kms audio path and force bcm2835 card 0 = hdmi card 1 = headphone jack
dtoverlay=vc4-kms-v3d-pi4,noaudio
disable_overscan=1
overscan_left=0
overscan_right=0
overscan_top=0
overscan_bottom=0

I tested this config with RaspiOS and it is working well.

On RPi4/Android11, kernel command line contains
video=HDMI-A-1:[email protected]

Framebuffer seems to be right set:
dmesg |grep -i framebuf
[ 0.323709] simple-framebuffer 3e6b4000.framebuffer: framebuffer at 0x3e6b4000, 0x546000 bytes, mapped to 0x(____ptrval____)
[ 0.323729] simple-framebuffer 3e6b4000.framebuffer: format=a8r8g8b8, mode=1920x720x32, linelength=7680
[ 0.323953] simple-framebuffer 3e6b4000.framebuffer: fb0: simplefb registered!

rpi4:/ # ls /sys/devices/platform/3e6b4000.framebuffer/of_node/
compatible format height name reg status stride width
conti_rpi4:/ # od -t x1 /sys/devices/platform/3e6b4000.framebuffer/of_node/height
0000000 00 00 02 d0
0000004
rpi4:/ # od -t x1 /sys/devices/platform/3e6b4000.framebuffer/of_node/width
0000000 00 00 07 80
0000004
rpi4:/ # echo $(( 0x780)) x $(( 0x2d0 ))
1920 x 720

logcat | grep -i surface
01-01 01:04:33.900 4786 4786 I SurfaceFlinger: Using HWComposer service: 'default'
01-01 01:04:33.901 4786 4786 I SurfaceFlinger: SurfaceFlinger is starting
01-01 01:04:33.910 4786 4786 I SurfaceFlinger: Disabling blur effects, they are not supported.
01-01 01:04:33.910 4786 4786 I SurfaceFlinger: Treble testing override: 'false'
01-01 01:04:33.912 4786 4786 W SurfaceFlinger: Unable to open libdolphin.so: dlopen failed: library "libdolphin.so" not found.
01-01 01:04:33.913 4786 4786 E SurfaceFlinger: Unable to open libframeextension.so: dlopen failed: library "libframeextension.so" not found.
01-01 01:04:33.942 4786 4786 I SurfaceFlinger: SurfaceFlinger's main thread ready to run. Initializing graphics H/W...
01-01 01:04:34.081 4786 4786 F SurfaceFlinger: Missing internal display after registering composer callback.
01-01 01:04:34.081 4786 4786 F libc : Fatal signal 6 (SIGABRT), code -1 (SI_QUEUE) in tid 4786 (surfaceflinger), pid 4786 (surfaceflinger)
01-01 01:04:34.164 4812 4812 F DEBUG : pid: 4786, tid: 4786, name: surfaceflinger >>> /system/bin/surfaceflinger <<<
01-01 01:04:34.191 4812 4812 F DEBUG : #03 pc 0000000000110ed8 /system/lib64/libsurfaceflinger.so (android::SurfaceFlinger::init()+3344) (BuildId: 2728199d2c528c4f4dbae73974da8516
)
01-01 01:04:34.191 4812 4812 F DEBUG : #04 pc 00000000000021a8 /system/bin/surfaceflinger (main+352) (BuildId: bb08dae93d29f9e39f6d8131b048a31e)

This is repeating every ~5 seconds

Any idea ?
Somethings to set into display config ?
I forget, nothing is displayed except the rainbow splash boot screen. Only a little part of services are started.
 

lumo76

New member
Feb 8, 2022
2
1
I have a similar issue as in #94 to get a "eGalax Inc. USB TouchController" working. The device is identified when plugin in:

Code:
rpi4:/ $ getevent -il /dev/input/event1                                      
add device 1: /dev/input/event1
  bus:      0003
  vendor    0eef
  product   0001
  version   0100
  name:     "eGalax Inc. USB TouchController"
  location: "usb-0000:01:00.0-1.1/input0"
  id:       ""
  version:  1.0.1
  events:
    KEY (0001): BTN_TOUCH          
    ABS (0003): ABS_X                 : value 1834, min 0, max 4095, fuzz 0, flat 0, resolution 0
                ABS_Y                 : value 2000, min 0, max 4095, fuzz 0, flat 0, resolution 0
                ABS_MT_SLOT           : value 0, min 0, max 0, fuzz 0, flat 0, resolution 0
                ABS_MT_POSITION_X     : value 0, min 0, max 4095, fuzz 0, flat 0, resolution 0
                ABS_MT_POSITION_Y     : value 0, min 0, max 4095, fuzz 0, flat 0, resolution 0
                ABS_MT_TRACKING_ID    : value 0, min 0, max 65535, fuzz 0, flat 0, resolution 0
  input props:
    INPUT_PROP_DIRECT

But it was not reaction correctly when taping on the screen. Thus I've copied the IDC file

Code:
cp Vendor_0483_Product_5750.idc  Vendor_0eef_Product_0001.idc

This was improving a lot, however it's fare from being precise. Is there a way to calibrate the touch screen or to improve the settings?

Thanks
 
Last edited:

Quantalou

Member
Jan 7, 2022
6
0
Hello,
We tried to generate the SDK with command make sdk and it failed.
How to build SDK ?

Regards
 

OldGodShen

Member
Feb 13, 2022
6
0
OmniROM

Download

Images taged with MICROG contain a pre-installed ready to use
microG setup https://microg.org/

Images tagged with WEEKLY can be used to setup
official Google gapps
  • 64bit build
  • Tablet mode configuration (not Android TV)
  • At least 4GB model recommened
  • No HW accelerated video playback
  • Mesa support for V3D is WIP so expect some graphic issues (especially on higher resolutions - HD 1920x1080 in general should be fine)
  • Support of booting from SD or USB by just changing config.txt (check README below)
  • GPIO support - monitor and set README.gpio
  • chrome (ium) is broken use alternative browser (firefox, edge)
  • for screen issues check https://www.raspberrypi.org/documentation/configuration/cmdline-txt.md
  • if you experience black bars connect to HDM1 or add disable_fw_kms_setup=1 to config.txt files
  • how to do updates? - check README.update
Adjust screen resolution:
To force a different screen resolution edit in /system/build.prop the line
#debug.drm.mode.force=1280x800

Remove the leading # comment and adjust to the screen resolution you want e.g.
debug.drm.mode.force=1280x720

Make sure to also check the README.txt in there
For questions please also use our official Telegram channel
https://t.me/OmniROM_Community

For current progress and how to build yourself check
README.build

XDA:DevDB Information
OmniROM, ROM for the Raspberry Pi 4

Contributors
maxwen, maxwen
Source Code: https://github.com/omnirom/android_kernel_brcm_arpi/

ROM OS Version: Android 11
ROM Firmware Required: 64bit
Based On: AOSP

Version Information
Status:
Stable

Created 2020-10-28
Last Updated 2020-10-28
How can I flash magisk?
 
Oct 17, 2010
34
6
copy the boot image to /sdcard and install the Magisk apk with adb install. When you click the install for Magisk (in the app), it will ask where he can find the boot image. He will create a patched boot.img to copy over the real boot.img. That is the best of my recall. If it doesn't work, post back here.
 

Top Liked Posts

  • There are no posts matching your filters.
  • 13
    OmniROM

    Download

    Images taged with MICROG contain a pre-installed ready to use
    microG setup https://microg.org/

    Images tagged with WEEKLY can be used to setup
    official Google gapps
    • 64bit build
    • Tablet mode configuration (not Android TV)
    • At least 4GB model recommened
    • No HW accelerated video playback
    • Mesa support for V3D is WIP so expect some graphic issues (especially on higher resolutions - HD 1920x1080 in general should be fine)
    • Support of booting from SD or USB by just changing config.txt (check README below)
    • GPIO support - monitor and set README.gpio
    • chrome (ium) is broken use alternative browser (firefox, edge)
    • for screen issues check https://www.raspberrypi.org/documentation/configuration/cmdline-txt.md
    • if you experience black bars connect to HDM1 or add disable_fw_kms_setup=1 to config.txt files
    • how to do updates? - check README.update
    Adjust screen resolution:
    To force a different screen resolution edit in /system/build.prop the line
    #debug.drm.mode.force=1280x800

    Remove the leading # comment and adjust to the screen resolution you want e.g.
    debug.drm.mode.force=1280x720

    Make sure to also check the README.txt in there
    For questions please also use our official Telegram channel
    https://t.me/OmniROM_Community

    For current progress and how to build yourself check
    README.build

    XDA:DevDB Information
    OmniROM, ROM for the Raspberry Pi 4

    Contributors
    maxwen, maxwen
    Source Code: https://github.com/omnirom/android_kernel_brcm_arpi/

    ROM OS Version: Android 11
    ROM Firmware Required: 64bit
    Based On: AOSP

    Version Information
    Status:
    Stable

    Created 2020-10-28
    Last Updated 2020-10-28
    2
    Builds
    https://dl.omnirom.org/tmp/rpi4/
    The plan is to have about weekly builds - at some point it will be part of all the other OmniROM devices with regular weekly builds
    once the Android 11 cycle starts

    20201016
    • Just a snapshot of the current state
    • Non microG builds will come later

    20201028
    • WEEKLY and MICROG build

    20201105
    • TWRP included
      IMPORTANT
      Check new recovery section in README for all the details

      EVEN MORE IMPORTANT as posted before to get a working TWRP I had to disable userdata encryption for the moment
      that means IF you try to boot this with your old userdata you will get an unpleasent surprise. You must start with
      a fresh userdata partition (mkfs + tune2fs). You CAN (if you know what you do) create an backup BEFORE
      updating by using adb shell and e.g. create a copy of userdata on an external USB device and copy that back afterwards.

      But after all of that hassle from now on updating will be much easier since with TWRP you can create
      userdata backups on an external USB device very easy

      So going from this release an update could work like this.
      1) boot your current install into TWRP
      2) create an userdata backup on an external USB device
      3) flash the new build - you dont need to care about userdata
      just do whatever you always do with it e.g. resize
      4) boot new build
      5) go into Advanced settings and switch boot mode to recovery and restart
      6) in TWRP restore the backup you created before from external USB device
      7) flash switch_boot_rom.zip and reboot system
      8) enjoy
    • add boot mode switch in Advanced settings
    • add audio output switch in Advanced settings
    • whatever has been done since the last build

    20201115
    • fixed 5GHz WiFi AP
    • fixed Fondar USB touch screen
    • actual development

    20201122
    • switch to new kernel base - big thanks to Konsta - with update to 5.4.77
    • update mesa to v20.2
    • initial work to support pi camera - not yet working but close
    • initial work to support gpio - IF you want to use gpio please contact me on tg cause I was too lazy to create a README for this yet
    • all core system updates and fixes

    20201128
    • performance tuning
    • November security update
    • gpio support update
    • all core system updates and fixes

    20201206
    • added gpioi-fan
    • cpufreq config in advancded settings
    • all core system updates and fixes

    20201214
    • property persist.debug.drm.mode.force to adjust screen resolution
    • all core system updates and fixes

    20201224
    • added disable_fw_kms_setup=1 to config.txt files. This should remove unneeded black bars for some screen when connected on HDM1. I did not see any problems on screens that have no issue with that anyway so I enabled by default. If you face issues remove line and report back please
    • enabled showing BT device battery level if applicable
    • all core system updates and fixes
    • Best Xmas wishes for those who care about it and looking forward to 2021

    20201231
    • add WiFi firmware for p400
    • kernel update 5.4.85
    • add userdata resize and project quota enable script in TWRP
    • reduce dB of max media volume
    • all core system updates and fixes
    2
    Is it possible to add magisk support?
    1
    Thank You very much for answer. I thing problem is with resolution. Touch matrix working well. Do you know, how to setup RPi4 boot up resolution for 800x480? Screen type is DSI-1
    https://www.alza.sk/raspberry-pi-touch-display-7-d4268133.htm

    Hello, i have the same problem. I would like to use the Omni ROM with my RPI4 an the 7" Raspberry Touch Screen. Please add the option to your nice ROM.
    Please
    1
    is there a changelog somewhere ? i see there is a lot of development at https://dl.omnirom.org/tmp/rpi4/ - but no changelog.
    I'd like to know if there is progress at USB-camera-support.