WINEDS

Bluetooth is fixed with AIC8800 wifi module.  Credit to @pocosparc  
 
 
Changed DTS Bluetooth node as per at least change suggestion.  @pocosparc did you confirm optical SPDIF as working?  My port is not illuminated.
 

Hi @Hqnicolas
 
I am currently still doing some research as I found out that Bluetooth does not work and it has to do something with the UART1 line. I will first try to fetch the low-hanging fruits.
Meanwhile, I replaced the debian with ubuntu image for now and I extracted all of the firmware drivers from my backup dump from the super.img file. I am attaching them to this post and they are coming from the vendor/firmware part. I currently do not know if they need to be statically or dynamically linked with the kernel. If they help in any way, we can include them in the armbian Linux. Link
 
I have a few questions for you as well. Do we have some sort of PCBA schematics and the HW setup with the external hardware: direct pins connections, what goes via muxes, etc.? This would speed things up. I requested this from the buyer, but the chances are really low that he will send anything back. 
What is the latest dts version? Do we have it in the form of mnemonics and not as bare pointers in byte values (&gpio instead of some 0x8e value)?
 
Now to the Bluetooth problem
I found somewhere in this forum that someone fixed the Bluetooth by changing the TX and RX pins in the dts. Link to git commit
 
Looking at what is currently in (I am using some rk3566-firefly-roc-pc.dts - there are multiple of them and changes are phandles for USB 2.0) and what is in the git and the version originally coming from this device show that there is a TX and RX swap in the file (maybe due to different PCB layouts and hoping that there is one bit that allows UART alternate function RX/TX swap - not sure as the datasheet for RK3566 is far from being useful).
 
So the file should be at least changed to:
bluetooth { compatible = "brcm,bcm43438-bt"; clocks = <0x8d 0x01>; clock-names = "lpo"; device-wake-gpios = <0x8e 0x10 0x00>; host-wake-gpios = <0x8e 0x11 0x00>; shutdown-gpios = <0x8e 0x0f 0x00>; pinctrl-names = "default"; pinctrl-0 = <0x8f 0x90 0x91>; vbat-supply = <0x24>; vddio-supply = <0x92>; status = "disabled"; };  
In the original dump file there is also UART8 messing around, but I am currently not sure what her role is. @Hqnicolas If you have everything setup on you side, can you quickly check this out? Or I will try to fiddle around with armbian-config and device tree entries and let's see where it gets me.

I think your driver is already in armbian, but your chip is not identified. To get it working you can identify it in the ethernet node this way:
 
mdio {
            compatible = "snps,dwmac-mdio";
            #address-cells = < 0x01 >;
            #size-cells = < 0x00 >;
            ethernet-phy@1 {
                compatible = "ethernet-phy-ieee802.3-c22"; ====> compatible = "ethernet-phy-id7b74.4411", "ethernet-phy-ieee802.3-c22";

https://kanyi.gm/how-to-install-and-enable-aic8800-chipset-wi-fi-driver-for-linux/
 
Some people have problems with this device
There’s no denying that the AIC8800 is a terrible component for a linux desktop/server. Its manufacturer has not made any effort to upstream the drivers. Either you live with no WLAN/BT or you live with DKMS package.

Thanks maka for replying.  Issue is resolved and Ethernet is working now.  I did not read the first email from Maxio tech properly and I was sent files for MAE0621A-Q3C rather than the MAE0621A-Q2C which I have.  Maxio tech kindly sent me files for MAE0621A-Q2C yesterday and I managed to compile it into the current kernel 6.12.24

The driver is well so the problem seems you are in the wrong ethernet phy. Change from phy0 to phy1 in the ethernet node.