All Activity

never heard of that one before. Serial console is enabled by default on all boards (lookup your board manual on where to connect) which also outputs uboot logs while on early boot stage. For increased kernel verbosity set verbositoy to 7 in /boot/armbianEnv.txt

Great！ Thanks a lot! I will try it to se if Diablo II can run tonight!

You can download dgVoodoo2 from here: https://community.pcgamingwiki.com/files/file/572-dgvoodoo-2/ (Unfortunately, newer versions of dgVoodoo2 won't work with Wine/Proton.) Extract the archive, then copy dgVoodoo2_79_1/MS/x86/d3d9.dll to the game's folder (just like Dxvk). Then, copy the Dxvk's d3d11.dll, dxgi.dll to the game's folder. If done right, you will see that the Dxvk HUD shows the game uses D3D11 API.

@chris178 I am currently working on a solution. Radxa typically uses rsetup and extlinux.conf to load overlays during boot; however, Armbian uses armbian-config along with boot.cmd and boot.scr. I have decided to use armbian-config to load the overlays instead.

http://blog.armbian.com/content/images/2026/01/githubhighlights-2.webpThis week’s Armbian development saw significant kernel upgrades across multiple platforms, including meson64, rockchip64, and UEFI, with several branches bumped to versions 6.18 and 6.19. Notable hardware support enhancements were introduced, such as SPI controller support for Allwinner A523, new board additions like mixtile-core3588e, and improved compatibility for devices including Radxa and Youyeetoo. The build system received stability and performance improvements, featuring parallel repository generation, refined CI scripts, and fixes for repository management in concurrent environments. Kernel configuration updates enabled additional filesystems and tickless idle for mvebu devices. Several patches addressed hardware initialization, network, and desktop issues, while deprecated packages were removed for a leaner build. Overall, the changelog reflects a strong focus on expanding hardware support, refining build infrastructure, and maintaining kernel currency. add allwinner: a523: Support SPI controllers patch series. by @EvilOlaf in armbian/build#9124allow build on forky host. by @amazingfate in armbian/build#9153CI: add watchdog action script and adjust log cleaning parameter. by @igorpecovnik in armbian/build#9147Fix repository management script for parallel execution. by @igorpecovnik in armbian/build#9163fix: replace comm with grep to support uutils coreutils. by @cantalupo555 in armbian/build#9148kernel configs - enable EROFS as module, ditto overlayfs & F2FS. by @tabrisnet in armbian/build#9135kernel: mvebu: enable tickless idle. by @neheb in armbian/build#9139meson64-6.19: de-bork PCIe, again (missed a patch from 6.18). by @rpardini in armbian/build#9151meson64: bump edge to 6.19; current 6.12 -> 6.18. by @EvilOlaf in armbian/build#9069meson64: bump edge to 6.19; current 6.12 -> 6.18 (nonsquash). by @rpardini in armbian/build#9146Missing variable in board config - address JSON matrix warnings. by @igorpecovnik in armbian/build#9144mixtile-core3588e: add board with vendor and edge branches (Joshua Riek + fixes). by @rpardini in armbian/build#9119Modify radxa-e54c board config to fix leds, network and Gnome desktop init.. by @schwar3kat in armbian/build#9111radxa-cubie-a5e: maint and fixes. by @EvilOlaf in armbian/build#9152Remove libfuse2t64 from package list. by @igorpecovnik in armbian/build#9155Repo tools: add parallel repository generation support. by @igorpecovnik in armbian/build#9150Revert few Rabbit's suggestions that broke repository management. by @igorpecovnik in armbian/build#9156rockchip64-6.18: Enable HDMI1 and audio for HDMI0/1 on CM3588-NAS. by @rpardini in armbian/build#9138rockchip64: add edge kernel support for Youyeetoo YY3588. by @SuperKali in armbian/build#9137rockchip64: fix broken patch in 6.18.3, rewrite everything. by @EvilOlaf in armbian/build#9157uefi-all: bump edge to 6.19 (except loong64); legacy 6.6 -> 6.12; current/cloud 6.12 -> 6.18. by @EvilOlaf in armbian/build#9086uefi-loong64-edge: bump to 6.19. by @amazingfate in armbian/build#9149View the full article

So the workaround is to set cpu freq max to 1 MHz. I still have to use a ghetto cooler 5000 when running a process that maxes out the CPU. Seems Duo running 10C+ at idle with latest kernel and Noble. I can live with that as long as I know. The default is 1.2 MHz max which runs too hot at 1.3V.

Java UIO Java UIO offers robust Java interfaces optimized for Linux Userspace IO, emphasizing high performance. Engineered from scratch, it leverages contemporary kernel APIs, libraries, and cutting-edge code generation methodologies, embodying a state-of-the-art cross-platform solution. Rather than reinventing established paradigms, Java UIO harnesses existing standards, avoiding redundant development commonly observed in other IO libraries. Seamlessly supporting JDK 25 LTS, this framework integrates Project Lombok to streamline code, enhancing readability and conciseness. 🌍 The 32-bit & FFM Reality With Java 25, there is no more support for X86_32, and ARM32 has limited JDKs available. While FFM (Foreign Function & Memory API) is the future, it is not supported on X86_32 or ARM32. For now, JavaUIO moves into the future providing a high-performance path for ARM32, while fully supporting ARM64 and X86_64.

Just to add, Armbian boots upto systemd and then gets stuck at loading some scripts. This image gives some additional information.

Hi, So, Previously this massage (No emmc) appeared if the preinstalled Android was able to boot up completely and then restarted by repowering while the sdcard was inserted or the tvbox was rebooted while the sdcard was in the slot, however this got fixed if the device was powered off during the android bootanimation but after accidentally installing an incomplete image of latest armbian release I am no longer able to get the emmc working even after trying restarting at various boot levels. I have found others experiencing the same issue in this post but according to the circumstances I am in I think my issue might have a different solution. Any way through which this issue can be solved ? This might help : root@multitool:~# dmesg -T | grep mmc2 [Tue Jan 6 21:06:28 2026] mmc_host mmc2: Bus speed (slot 0) = 2343750Hz (slot req 400000Hz, actual 390625HZ div = 3) [Tue Jan 6 21:06:28 2026] mmc_host mmc2: Bus speed (slot 0) = 50000000Hz (slot req 50000000Hz, actual 50000000HZ div = 0) [Tue Jan 6 21:06:28 2026] mmc2: Card stuck in programming state! __mmc_switch [Tue Jan 6 21:06:28 2026] mmc2: error -110 whilst initialising MMC card I have also attached the whole dmesg log and lsblk output if needed. multitool-ssh-output.txt

Hi all, Is there any working way to enable detailed logs from uboot? I am using fresh main branch (latest commit is 4648add1450c1b26bb29bf412ef054b0e978947d) . I have created following config inside userpatches/config-test.conf.sh file BOARD="orangepizero" BRANCH="current" RELEASE="jammy" BUILD_MINIMAL="yes" PREFER_DOCKER="yes" DOCKER_ARMBIAN_BASE_IMAGE="ubuntu:jammy" BOOTDELAY=2 UBOOT_DEBUGGING="yes" My command to build image is ./compile test. As I understand from source code UBOOT_DEBUGGING variable should enable such detailed uboot logs, but it seems like build framework ignored this option, I did not get detailed logs from uboot, also build logs do not contain info messages related with enabled UBOOT_DEBUGGING, etc. P.S. Just for context - I have board wukongpih3zero (clone of orangepi zeor) and booting is hanging at unexpected mcc1 slot instead of microsd mcc0 - that's why I need detailed boot logs.

I guess the image was built from a pending pull request for testing before merge. Since it has been merged now it should work. Make sure your framework clone is up to date (git pull).

@Werner can you please tell how did you built this image: https://testing.armbian.de/Armbian-unofficial_26.02.0-trunk_Radxa-cubie-a5e_trixie_edge_6.18.2_minimal.img.xz ? Image I get with: ./compile.sh build BOARD=radxa-cubie-a5e BRANCH=edge BUILD_DESKTOP=no BUILD_MINIMAL=yes KERNEL_CONFIGURE=no RELEASE=trixie seems to be far away from your result. First and foremost it does not boot 😉 I ask because of two reasons: - I'd love to be able to build myself an image that I'm running. - I need to be able to mount ZFS volumes. I'm unable to install `zfs-dkms` and `zfsutils-linux` when using your image. `apt` says that kernels newer than 6.14 are not supported. I'd also appreciate any tips or resources that would help me to get 6.12.x (or whatever lower than 6.14). I experimented a bit but without luck. I saw that it seems that `current` branch should not be supported on radxa-cubie-a5e board but I somehow got an image that at least has 6.12.63 in its name. It doesn't bring much because it does not boot.

Oh, I understand. Camera, hardware video decoder and hardware video encoder - they all appear now as /dev/videoX devices, because new hardware videocodec accelerator driver (or standalone video decoder and video encoder drivers) became V4L2 kind of driver. Also thanks for the reference link.

The ffmpeg conversation in https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20847 suddenly stopped 2 weeks ago. Does it seem like there's a big obstacle suddenly, or just a well deserved engineer vacation going on?

How did you apply dgVoodoo2 ? I download the archieve file it is a bunch of dll.

Thank you for testing my instructions I was on vacation, and I didn't remember to check this until now Which LCD are you using? Which kernel driver are you using for it? All the questions in your post depend on the LCD model and the driver you used. (I am using panel-mipi-dbi, which I know that it has lost its ability to display X11 graphics since a few months ago) Can you 100% say that you don't have X11? You don't need to use the same kernel version I use. The newer version you have is preferable 🆗 I am glad you can use the Trixie labwc. Personally, I am using Bookworm in the meantime, since its the only way I know to get H264 hardware acceleration. With Trixie's labwc, do you ever get desktop crashes when you log out?

I updated my patches to the new style (using series.conf) and that took some time. They are back online, as are some new images. I post more updates at the opi discord btw. As far as I know, neither network, spi flash or pcie are working currently.

Current way of installing is as written on the download pages. It was verified with latest v24.11.2 image.

I'll try to explain in some details. 1. Yes these are different things even if they appear all as /dev/videoX devices. In case of camera, hardware video decoder and hardware video encoder one may have /dev/video0, /dev/video1 and /dev/video2 2. no, because really the DRIVER does the encoding or decoding provided there is codec hardware present. The driver provides the userland communication to the hardware. This happens through the v4l2 layer (Video For Linux 2) which does the abstraction for different types of hardware, eg. H3, Hantro, Intel ... Of course this requires a v4l2 kind of driver, whereas the cedar_dev (aka cedar_ve) is NOT. 3. Right, bootlin is a little more 🙂 I mentioned these because they provide the latest cedrus driver with encoding support (with still some bugs...) 4. yes, cedar IS outdated. The most modern concept is v4l2 (see above). Cedar and this specific ffmpeg fork are using a different approach. 5. I never mananged to get this combination to work, at least not on the latest kernel. HTH T. Reference: https://bootlin.com/blog/open-source-linux-kernel-support-for-the-allwinner-v3-v3s-s3-h-264-video-encoder/

when send command poweroff or reboot alywas display client_loop: send disconnect: Broken pipe instead in on ArmBian Debian 12: Connection to 192.168.1.12 closed by remote host. Connection to 192.168.1.12 closed. but maybe on Debian 13 now is different message after reboot or poweroff

# cat /boot/armbianEnv.txt verbosity=1 bootlogo=true console=both extraargs=cma=256M overlay_prefix=rockchip-rk3568-hk fdtfile=rockchip/rk3568-odroid-m1.dtb rootdev=UUID=0a9f09a7-bbb5-493f-a7af-b0e62f4aff8e rootfstype=ext4 usbstoragequirks=0x2537:0x1066:u,0x2537:0x1068:u It seems the UUID can be changed in `/boot/armbianEnv.txt`.

@Ducdanh Nguyen, virtually every adapter will work. I'm using DFRobot multiplexer v.1.0 (FT232R-based): https://wiki.dfrobot.com/Multi_USB_RS232_RS485_TTL_Converter__SKU_TEL0070_

During installation new Image show on screen following problems: Failed to retrieve unit state: Transport endpoint is not connected _ _ _ _ _ /_\ _ _ _ __ | |__(_)__ _ _ _ __ ___ _ __ _ __ _ _ _ _ (_) |_ _ _ / _ \| '_| ' \| '_ \ / _` | ' \ / _/ _ \ ' \| ' \ || | ' \| | _| || | /_/ \_\_| |_|_|_|_.__/_\__,_|_||_|_\__\___/_|_|_|_|_|_\_,_|_||_|_|\__|\_, | |___| |__/ v26.2 rolling for Orange Pi Zero running Armbian Linux 6.12.63-current-sunxi Packages: Debian stable (trixie) Support: for advanced users (rolling release) Failed to retrieve unit state: Transport endpoint is not connected trying reboot: root@orangepizero:~# reboot Call to Reboot failed: Connection timed out

I found a way to install 25.04 onto emmc with no problems afterwards. Updates and upgrades just fine now. https://pastebin.com/DzNYtmQa

Looks like stuff got lost due to force-push. Check this how it looked before: https://github.com/rvdr/build/activity?ref=spi-nvme-patches