Unfortunately, Link Star H68K is not one of the supported boards. Hinlink H68K is Community Maintained and is provided as-is with no guaranteed support and no any maintainer listed. If you can figure out what is different in the bootloader and device tree, you should be able to adapt what is in the Hinlink H68K build configuration.

Or probably the easier way is to edit /etc/default/armbian-zram-config

@Biasio95 Wait, I'm an idiot, the kernel parameter disables zSWAP, not zram. sry. Try "swapoff /dev/zram0" (or whatever the dev is named) and then remove the module with "modprobe -r zram" "echo 'blacklist zram' >> /etc/modprobe.d/blacklist.conf", add another swap and reboot. Or you could disable the zram-generator with the kernel parameter "systemd.zram=0", that way no zram should be able to be created. One of the methods should work, but I have not actually tried it on armbian, so there might be configs that can throw errors if you just remove the module, IDK. If not, read: https://docs.kernel.org/admin-guide/blockdev/zram.html#add-remove-zram-devices and remove the device in /sys/class/zram-control/hot_remove, I would assume that would stick. I assume you know how to setup a "normal" swap partition, because running without swap altogether is prob not a good idea.

I'm getting the same serial output stopping in the same place on my board: https://github.com/yisding/rock-5b-ysp/blob/main/findings/2026-07-09-rock5b-armbian-sd-boot-investigation.md Both zeroed out SPI and working (26.5.1) SPI don't boot. Got it to spin up HDMI by dd-ing over some new 26.5.1 bits over to the SD card but still not quite working. Good news is that 26.5.1 Ubuntu 26.04 does boot, and it also has a package for the 6.1 vendor kernel which boots also. I'm sure there's going to be some kernel/userspace mismatch between 6.1 and 26.04 so I've been "vibe-porting" the hardware encoding/decoding drivers to 6.18. Not quite ready yet for distribution, although it does work with ffmpeg and makes gnome-remote-desktop run silky smooth.

Just managed to distract myself some more in this outrageous heat here in the UK and can confirm that the RELEASE=stable (6.18) build (from compile.sh) boots ok after some tweaking. It got stuck on "Starting Kernel" but when I added these lines to /boot/armbianEnv.txt to see why it failed, then it booted ok - dunno why. I think keep_bootcon was the important bit. extraargs=earlycon=uart8250,mmio32,0x02500000 keep_bootcon ignore_loglevel loglevel=8 The other problem then was getting the ethernet to work which I did by creating a device tree overlay file - much the same as other here is it only seems be working at 100M unlike the 1000M in the 5.15 kernel. The extra problem seemed to be that the wifi driver would periodically crash and cause a kernel-panic. The following DTS file pulls the ethernet down to 100M and disables the wifi to stop it crashing. You then compile this into a .dtbo file, copy it to /boot/overlay-user and reference it in armbianEnv.txt with this line user_overlays=opi4a-network-fix The opi4a-network-fix.dts syntax to create it is as follows ... apart from that I can't get it to do much more, so will go down the 5.15 bookworm route until things change /dts-v1/; /plugin/; / { compatible = "xunlong,orangepi-4a", "allwinner,sun55i-t527"; /* Maintain working Ethernet at 100M address 1 */ fragment@0 { target-path = "/soc/ethernet@4510000"; __overlay__ { phy-mode = "rgmii-id"; phy-handle = <&ext_rgmii_phy>; status = "okay"; mdio { #address-cells = <1>; #size-cells = <0>; ext_rgmii_phy: ethernet-phy@1 { compatible = "ethernet-phy-ieee802.3-c22"; reg = <1>; max-speed = <100>; }; }; }; }; /* Hard disable the unstable wireless MMC controller block */ fragment@1 { target-path = "/soc/mmc@4021000"; __overlay__ { status = "disabled"; }; }; };