I successfully flashed the eeprom to boot 0xf46 but I still get no response from the display, I plugged in an HDMI monitor and was able to capture this info. It seems like it’s only seeing the eMMC disk though. I flashed the NVME with Rex’s Kali version. But no luck yet. (hackerGadgets CM5 adapter board + NVMe battery board upgrade )
The red activity light stays illuminated, but the green activity light on the NVME battery board only flashes a few times then stops.
I enabled PCIex1=on in /boot/config.txt
SMH
@Rex This just caught my eye. In your example in the very first message in this thread …
# Default BOOT_ORDER for provisioning
# SD -> NVMe -> USB -> Network
BOOT_ORDER=0xf461
The comment says SD -> NVMe -> USB -> Network, but the hex value doesn’t have the 2 in there, so it would never try to boot from the network.
Should the value in the example be 0xf2461, or should the comment not include the -> Network at the end?
For the benefit of others who may be curious, this page documents the BOOT_ORDER values.
Things are not going as well for me…..
vcgencmd version and sudo vcgencmd version both give me :
VCHI initialization failed
and rpi-eeprom-update gives me:
rpi-eeprom-update: command not found
Im using Rex’s Kali ISO on an SD card.
The “VCHI initialization failed” error on a Raspberry Pi typically occurs when the user does not have the necessary permissions to access the VideoCore hardware. To fix this, you need to add your user to the “video” group using the command sudo usermod -G video your_username and then restart your system or the relevant service.
Thank you, But that is NOT it.
It’s a fresh install of Rex’s Kali ISO. So Kali is the only user, I used sudo usermod -aG video kali and no bueno.
I had found this previously, but in a different thread. Thank you for pointing me to this one, it was a better explanation, but still not the fix.
EDIT : Using ID, I can confirm that kali is in fact in the video group. 44(video)
Try doing it with Trixie to update and edit the eeprom
Will do, is it persistent on the CM5 ? or still on the SD?
Correct it will stay with the CM.
That did it! Thanks again Rex!
Take out that waveshare spacer, make a lot of troubles special with CM5.
Thanks, but.. This was already solved months ago, and there is no spacer used 
Its just a CM5 with the official adapter and official mainboard. At least it was at that time 
Where have you heard that ??
I am using a CM5, without a spacer. I do NOT have any heat issues, I do however have a heat skink and fan that I will be installing once I get a new back plate printed.
Maybe its the SD card or I’m doing something wrong..
I’ve done the EEPROM update as linked by Rex and saved the config: It looks like the config isn’t saved.
I can only boot up images freshly written wit TwisterOS, Trixie and DragonOs (others don’t work) and after reboot I get kernel panic on the external monitor..
I have a CM5 lite 8G.
You’re having a different error then one associated with the eeprom. Try putting kernel=kernel8.img in you config.txt in the pi5 section. I noticed building a new Trixie image that the 16k kernel isn’t booting using kernel8 will make the cm5 use the 4k kernel.
I’ll try to do (kernel8) that.
For now, I can only boot up DragonOS and have done the eeprom-update & after that used raspi-config when booted.
In raspi-config, I’ve made changes the NVME speed to 2x and reviewed the boot order (SD/USB/NVME/Network) and instead of reboot I did shut down.
Rebooting doesn’t work (hangs/crashes), but shutdown and startup does work now.
I’ve ordered the HG upgrade kit without extension cards (I have the AIO v1) and requested to modify the NVME-battery board with a reset/power switch: That will greatly help the hanging issues without taking batteries out 
did you not only update the eeprom but do the edits i have outlined here.
I would also not put usb before sd and nvme or else it with slow down booting quite a bit.
Hi Rex,
It’s been a while and I did the advised configs. I can now run Bookworm, Trixi and DragonOS but still without reboot. Reboot lets the startup hanged,
The GPS, TAR1090 and SDR++ brown works and only Meshtastic (even after multiple reinstalls) refuses to start.
My bootconfig:
My Meshtastic yaml-file:
I also checked my Meshtasticd service:
If I check the status, It still gives me this problem:
I wish we had a command to check if the Lora-chip is working or not, because I can’t find it in lsusb:
Hi Rex,
It’s been a while and I did the advised configs. I can now run Bookworm, Trixi and DragonOS but still without reboot. Reboot lets the startup hanged,
The GPS, TAR1090 and SDR++ brown works and only Meshtastic (even after multiple reinstalls) refuses to start.
My bootconfig:
My Meshtastic yaml-file:
I also checked my Meshtasticd service:
If I check the status, It still gives me this problem:
I wish we had a command to check if the Lora-chip is working or not, because I can’t find it in lsusb:
Oh, there is a program in System Tools called USBView, herein I got the UC AIO Ext marked in red: Is that normal?
