Ok ,what is AH2 short for?I don’t think all addon authors/builders/skinners have migrated to Piers22,
Try the new dump EDID function of CE-NO.
1 Like
I get the same green stuck screen on my SC2 device, but it happens only if using suspend mode then restart/reboot to emmc.
I just use it for CE and I rarely do any restarts this way for Updates. So It’s not a real big issue for me.
Edit: I tried the EDID dump option under CE/HW settings and it’s no different, still shows green screen when suspend & restart/reboot to EMMC is utilized.
1 Like
i have installed ce-no with single boot use on emmc, also tried to wol with dual boot (android 11/14 and ce usb stick) neither worked
@Portisch - I downloaded nightly 20251207 and was doing some more extensive audio testing before I burn to eMMC. In Output Configuration mode “Best Match” and “Optimised” all is good and working in accordance with your changes. Everything is ok including reboots.
However, in Output Configuration “Fixed”, it switches to the set speaker layout and sampling rate ok, but after a reboot the output reverts to 2.0 and 48kHz. Every channel (except LFE) is mapped back correctly to the two front speakers, but the “Fixed” setting is not obeyed until you change the setting in Kodi and then it works again until the next reboot. It would seem that when the code sees the “Fixed” setting on initial start that it doesn’t evaluate the user setting and keeps the default 2.0 48kHz until after the user changes them again.
Do you need a log file from me or is the above info enough?
Good afternoon, here is the file after the screen turns green after restarting Tox3! When moving the controller’s pointers I can hear the navigation sounds normally, as if I were navigating through the menus. However, the screen is green.
edid.bin (513 Bytes)
This function is not to dump for debug. It’s to cheat your system a good EDID caused bad by an AVR.
1 Like
The AVR/Sounbar isn’t the issue, cause I get same result W/O using Q930F connected to eArc (LG OLED). The SC2 box is directly connected to TV. But I’m totally fine with using solo CE/KODI22
1 Like
I understand, it’s a hack to see if it solves the problem, but without success. It’s something related to the new kernel, because in CE NE 22 December 2024 this doesn’t happen, or in CE 21… Let’s see if more people show up with the problem on the S905x4, but I appreciate the help!!! Have a good week.
Edit: I restarted on Android and then restarted back to CE, and it seems to have stopped giving the green screen when restarting CE… if it returns, I’ll edit here again. Thank you.
1 Like
No log from a problem is no problem.
That addressed to me? I thought you would be familiar with your changes over the past few days!
Can you check the problem with the Ugoos AM8? As I wrote above, after updating the kernel to 5.15.170, the device no longer recognizes any connected USB devices. Sometimes you have to restart several times for them to be recognized, sometimes that doesn’t help. The connected devices are a WD hard drive with its own power supply and a Logitech Unifying receiver.
As I said, I had no problems with the old kernel.
1 Like
I pushed the command into config.ini, but it still doesn’t work. The WD hard drive is not recognized, no problem when starting Android, and there was no problem with the old kernel either.
When you view cmdline:
cat /proc/cmdline
do you see “usbcore.autosuspend=-1” there?
1 Like
Yes, it is displayed.
Please paste the result of lsusb
CoreELEC:~ # cat /proc/cmdline
systemd.show_status=auto init=/init loglevel=${loglevel} console=ttyS0,921600 no _console_suspend earlycon=aml-uart,0xfe078000 loop.max_part=4 scramble_reg=0xfe0 2e030 otg_device=1 ug_magisk=0 logo=osd0,loaded,0x3e800000 vout=2160p60hz,enable hdmitx=,420,10bit hdmimode=2160p50hz hdmichecksum=0xf5ed0000 hdr_priority=26843 5472 frac_rate_policy=1 hdmi_read_edid=1 cvbsmode=576cvbs cpufreq.default_govern or=performance osd_reverse=0 video_reverse=0 kvm-arm.mode=none usb3_off=0 config _csc_en=1 mac=90:0e:b3:fd:4d:54 BOOT_IMAGE=kernel.img boot=LABEL=COREELEC disk=L ABEL=STORAGE console=tty0 keymap=us remotewakeup=0xffffffff decode_type=0 r emotewakeupmask=0xffffffff vout=2160p50hz,dis frac_rate_policy=0 hdmitx=, hdr_ policy=1 quiet reboot_mode=shutdown_reboot androidboot.bootreason=shutdown_reb oot androidboot.selinux=permissive androidboot.firstboot=1 androidboot.bootloade r=01.01.251203.121930 androidboot.hardware=amlogic androidboot.production_mode=0 androidboot.ddr_size=4096MB androidboot. androidboot.wificountrycode=US android boot.serialno=AM8230800000000118 androidboot.rpmb_state=0x1 androidboot.verified bootstate=orange usbcore.autosuspend=-1
CoreELEC:~ # lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 004 Device 002: ID 2109:0817 VIA Labs, Inc. USB3.0 Hub
The “TurboSight TBS 5930 DVB-S/S2/S2x” tuner is plugged into the USB port but is not recognized.
After restarting a few times, it will recognize the tuner and flash drive.
CoreELEC:~ # lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 002: ID 2109:2817 VIA Labs, Inc. USB2.0 Hub
Bus 003 Device 003: ID 734c:5930 TBS Technologies China TBS 5930
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 006 Device 002: ID 2109:0817 VIA Labs, Inc. USB3.0 Hub
Bus 006 Device 003: ID 346d:5678 Dahua U106
X96 M200 (S905X5M / S905A) - CE-NO 22-20251207
Issue: message “missing parition ‘super’ on eMMC” (boot from SD card)
Greetings 
I have X96 M200 device purchased mid-November, as I saw CE-NO alpha status and I wanted to help advance the development, and also I saw several members in the discussion having success with X96 M200 / M300 around August and September this year.
With my device it is different story for now. First with CE-NO 22.0-alpha2-generic, I had the ‘no super partition’ message. I was very confuse because of the success of other members, and I remember seeing one ‘super’ partition during boot. As I am new to CoreELEC I did research and learning a lot, found the very elegant solution of systemd / Android partitions mount by CoreELEC team 
, but I was still scratching the head for many days because of EROFS mount issue for my device (dmesg alpha2 log for the device).
Now with Nightly 20251207, EROFS issue is gone thanks to @vpeter and ef3c3c…90 commit, thank you very much for this revert! 
system and vendor partitions now mount very good. (dmesg nightly-20251207 log for the device).
However, after the partitions mount, I understand tee-loader script is not able to (pre)load with success the video_ucode.bin firmware. Unfortunately at this stage it is beyond my current knowledge and understanding, and any help and knowledge will be very much appreciated to move forward.
My hypothesis: software is different (maybe hardware?) from the X96 M200 / M300 devices in previous discussions here in August and September.
On my device: u-boot indicates 2025-09-09 version, and factory software build indicates 2025-09-25, therefore after the software for devices working in August and September even with CE-NO 22.0-alpha1. Kernel is same 5.15.170.
Version printed on PCB indicates BM972_262_V1.1 24313, in case. Also the S905X5M looks to renamed to S905A, this is a bit confuse for me if exactly same or no. Apparently yes, but not 100% clear.
If any useful, here is ‘super’ partition image dump from Android shell, and following is UART log from Android boot after factory reset (I found UART log is more verbose after factory reset than regular boot).
One last item to maybe help progress CE-NO in global: when investigating my X96 M200 issues, I did notice tee-loader show ucode version is 97.105, which after reading CoreELEC forum is very weird for me, as I understand version is around 4.something, right?
# /lib/coreelec/tee-loader start
run tee from android start
fw active slot: '_a'
active slot: '_a'
Android ucode version: '97.105'
run tee from android end
tee from android match SCS version, trying from coreelec
run tee from coreelec start
Using CoreELEC ucode file '97.105' for S905X5M
Using real rpmb for tee-supplicant
fw_path = /lib/firmware/video/video_ucode.bin
ree skip header length 0x8000
MM-module-name:VDEC CA,Version:1.0.21-g006fb97(build:487)
TEEC_Opensession failed with code 0xffff000f origin 0x3
tee preload video fw fail
run tee from coreelec end
using tee from coreelec failed
cleanup tee start
/lib/coreelec/tee-loader: line 147: 1271 Killed tee-supplicant
cleanup tee end
If correct, then apparently several issues in /usr/lib/coreelec/read-firmware-version, most important:
- line 19:
breakinifconstruct has no effect, only in loop (ex:while,do,until) - line 14:
[[ ... ]]construct is not POSIX, and thus work good in Bash, but undefined behavior inshwith weird results - line 12:
hedxumpstring conversion has edge case, because of non-print ASCII characters, was not easy to find, but as result wrong magic is identified, thenifcondition fails so the offset is modified and result in version 97 ucode instead version 4 ucode.
## execute on CE-NO 20251207 on X96 M200
## copy to add read_firmware_version call, no other change
# sh -x /tmp/read-firmware-version
+ read_firmware_version /lib/firmware/video/video_ucode.bin
+ local firmware_file=/lib/firmware/video/video_ucode.bin
+ minor=0
+ batch=0
+ '[' -f /lib/firmware/video/video_ucode.bin ']'
+ local offset=0
++ rev
+++ hexdump -e '1/4 "%s"' -n 4 -s 0 /lib/firmware/video/video_ucode.bin
++ echo $'PWEN\177'
+ magic=$'\177NEWP'
+ [[ NEWP != \N\E\W\P ]]
+ [[ NEWP != \P\A\C\K ]]
+ offset=256
++ rev
+++ hexdump -e '1/4 "%s"' -n 4 -s 256 /lib/firmware/video/video_ucode.bin
++ echo $'s7d_\177'
+ magic=$'\177_d7s'
+ [[ _d7s != \P\A\C\K ]]
+ break
++ hexdump -e '"%d"' -n 1 -s 272 /lib/firmware/video/video_ucode.bin
+ minor=97
++ hexdump -e '"%d"' -n 1 -s 276 /lib/firmware/video/video_ucode.bin
+ batch=105
+ echo 'minor="${minor}"; batch="${batch}"'
minor="${minor}"; batch="${batch}"
I hope this is of help for CE-NO! 
And I also thank in advance for all help and knowledge to solve my X96 M200 issues.
EDIT: corrected URL for dmesg nightly-20251207 log
After reboot
journalctl -l | paste
And upload your /lib/firmware/video/video_ucode.bin file somewhere just in case.
And I think it would be easier if you would enable terminal sharing for faster fix: coreelec:term_share [CoreELEC Wiki]