bolet75

@dma I'm using Ubuntu 22.04 with 6.1.43 vendor kernel, and am fine with mostly everything (wifi, bt, screen). Updating to more recent kernels (including 6.1.75 vendor) was always breaking something (screen resolution over 1080p was one of them), so I froze kernel updates. Vendor images lack specific Armbian improvements I got used to love (removed snaps, added zram swap and ramlog), so I'm not considering them. The update to Ubuntu 24.04.1 stopped halfway, and my system was unbootable (a btrfs snapshot was great for restoring everything). Will try 24.04.2 upgrade when it's there. I keep checking new Armbian Ubuntu desktop images from time to time. One path would be to start fresh with a new Ubuntu 24.04 image, downgrade the kernel if needed, and restore my private stuff (configs + apts). Quite cumbersome. I'm aware that Armbian has limited resources, am grateful for their work supporting so many SBCs: I have it running smoothly as server on 6 different models.

Great news! Any hope to have both on-board and pci-e cards recognised as they used to be in 5.10.160-legacy?

See this thread, it is possible to add the needed driver from dkms, then WiFi somehow works:

Success! The repository is at https://github.com/lwfinger/rtw8852be and has detailed instructions to compile and install. I followed the instructions and have WiFi working again on kernel 6.1.43-vendor-rk35xx The only difficulty was to find the correct headers package: linux-headers-vendor-rk35xx My OPi5Plus has the additional WiFi PCIe card, which was visible with the previous vendor kernel version, 5.10.160-legacy-rk35xx or package linux-headers-legacy-rk35xx. Though managing both cards used to be buggy, I only see one card now. Not perfect but enough for my current needs.

The problem is gone with another monitor, still on HDMI connection. As I'll keep that new one, I won't be able to test this issue in the future. As I said early, the issue is solved in newer kernels (not yet in 6.1.43 though which I upgraded to before I switched the monitor). So the issue looks like depending on both kernel version and monitor model. The fix which happened in recent kernels probably isn't worth backporting.

Oh thanks a lot! I could probably use the 1st patch to the armbianEnv.txt file: the root volume is on a usb drive, so the 2nd option would not work for me. But I still believe there is some hardware issue because even when USB ports work, the voltage supplied to is far below 5V. And the issue I have has been getting worse over the years, as if a component was wearing out. I'll report back when I can give it a try.

The USB power has been a growing issue with my C2 for years now. To have the board use the SSD through a USB/SATA adapter, I now have to power the adapter (which has a Y cable) from a neighbour board. It used to work long time ago without the other board "help". So there is probably a weakening component on my C2. Right now the voltage drops between the +5V barrel alim centre, and USB connectors' 5V. I see no rationale in this. My C2 is now decommissioned (I bought a OPi 3 LTS instead). So I will try to to solder a wire under the board to have the incoming 5V straight to the USB ports. It should not be worse than the Y USB cable bringing the 5V from the neighbour board.

Odroid-C2 (there was no adequate tag) After I upgraded my ssd install from 23.11.1 to 24.2.1, the board didn't boot. When starting 24.2.1 from a fresh sd card, usb and disk leds are on, and the board boots. But they don't show in lsusb. If I reconnect them after the board is booted, the leds stay off (and obviously they don't show in lsusb). Of course, booting from external ssd stops at some point, because usb is lost during the boot (when connecting a display, the failures start to happen on armbian services). Had to restore previous release 23.11.1 from backup on my ssd. Everything works again.

Previous images had a smaller gap before partition 1. My guess is that this space is now there to allow a future switch to UEFI boot, which as far as I understood, needs an EFI partition to be 1st on the disk, and it needs that much space. Someone here probably knows better than me. I already tried this new layout, with EFI on partition 1, and "generic arm64" kernel on partition 2. No success yet. Will dig into this again some time soon.

Contrary to 5.1.160, the sound output menu is empty when booting with 6.7.0-rc4 I was considering a DVI adapter to stay "digital". But your solution is bolder! The overscan doesn't happen on another 1080p hdmi monitor. I looked in the edid informations which appear in: /sys/devices/platform/display-subsystem/drm/card0/*/edid I found "wxedid" to decode and edit the edid (it's a tricky binary format). The relevant "timing" informations are identical between the two monitors. I tried editing the edid and writing it back: no change. Nor does it seem to change anything when given as a boot parameter.

The issue seems solved in kernel 6.7.0-rc4. But other issues appeared (among which no more sound). For now I'm staying with 5.1.160.

Using armbian-install I achieved to have my board boot with no sd card inserted. Try the options which use the "mtd flash" (mtdblock0 below) lsblk output: NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS mtdblock0 31:0 0 16M 0 disk mmcblk0 179:0 0 233G 0 disk ├─mmcblk0p1 179:1 0 512M 0 part /boot │ /media/mmcboot └─mmcblk0p2 179:2 0 232.4G 0 part /media/mmcroot mmcblk0boot0 179:32 0 4M 1 disk mmcblk0boot1 179:64 0 4M 1 disk zram0 254:0 0 7.7G 0 disk [SWAP] zram1 254:1 0 50M 0 disk /var/log nvme0n1 259:0 0 931.5G 0 disk ├─nvme0n1p1 259:1 0 64G 0 part │ └─cryptswap 253:0 0 64G 0 crypt [SWAP] ├─nvme0n1p2 259:2 0 64G 0 part /var/log.hdd │ / └─nvme0n1p3 259:3 0 803.5G 0 part └─home 253:1 0 803.5G 0 crypt /home mtdblock0 is a small and slow rom which is on the board itself (not on the emmc module), contains the bootloader and is not mounted. Maybe this is what's missing to have your board boot without sd card.

When connected to some PC monitors, the image is larger than the physical screen: about 20 pixels at the top and bottom sides, and 50 pixels at the right and left sides are missing. This happens as soon as the board boots (without logo), the start of lines are truncated, and this also happens on text ttys (Ctrl + Alt + F1..6). I guess it's about a boot env variable, but I could not find it.