All Activity

Vendor kernel will stay for awhile (this year for sure) as mainline is not on this level yet. This part might not be part of the release - you can grab it from nightly release - but most likely it won't have any affect to this topic.

ok, i guess i just dd new image then. have you read github release a few minutes ago? orangepi5: bump u-boot to v2026.04-rc3

Welcome to the latest Armbian Newsletter, your source for updates, community highlights, and behind-the-scenes news from the world of open-source ARM and RISC-V computing. Streamlining the ecosystem This past quarter development has taken a significant step forward for Armbian. We've been busy cleaning up and optimizing our codebase, merging over 150 improvements primarily focused on advancing kernel support for Rockchip and Allwinner platforms. A significant secondary effort was dedicated to modernizing the project's infrastructure and build tools, driven largely by a core group of dedicated contributors. Hardware support continues to expand. We're thrilled to introduce new LTS kernel v6.18 based images and support for exciting new boards like the SpacemiT MusePi Pro, Radxa Rock 4D, Orangepi RV2, Odroid M2, ... We've also brought back KDE Neon desktop builds and added RISC-V XFCE desktop support for those exploring new territory. Our user tools have received major upgrades too. Armbian Imager now features faster decompression, enhanced security with code signing for macOS and Windows, AI-powered translations, and a new settings panel with developer options. Behind the scenes, we've strengthened our build infrastructure to keep everything running reliably as we grow. Whether you're trying the new stable kernel or experimenting with our latest builds, this update reinforces our commitment to delivering a professional, accessible Linux experience for ARM and RISC-V hardware. Join Armbian at Embedded World 2026. Meet us in Hall 3, Booth 3-556 (Seeed Studio), where we’ll be showcasing the Armbian build framework and how it powers reliable, production-ready Linux for ARM devices. Download the latest release and experience Armbian today. SPONSORED Join us in making open source better! Every donation helps Armbian improve security, performance, and reliability — so everyone can enjoy a solid foundation for their devices. Release v26.2.1 · armbian/buildChanges sunxi: refactor full patchset. by @EvilOlaf in armbian/build#9219 “get completely rid of dead code toolchain stuff”, pt2. by @rpardini in armbian/build#9218 “get completely rid of dead cod…GitHubarmbianForget third-party utilities: meet Armbian ImagerArmbian Imager eliminates the guesswork from flashing SBC images. Real-time board detection, persistent caching, and built-in safety make installation fast, simple, and risk-freeArmbian blogDaniele BriguglioArmbian 2025: by the numbersOpen hardware is growing faster than ever and breaking in new ways. 2025 has been a productive year for the Armbian project. As the Single Board Computer ecosystem continues to fragment and expand, Armbian has consolidated its position as the universal glue holding the open-source hardware world together. Our missionArmbian blogMichael RobinsonThe Evolution of SBCsThe Evolution of SBCs: From Hobby Boards to Edge Computing Over the past two decades, single-board computers (SBCs) have transformed from experimental maker tools into the backbone of modern embedded and edge systems. What started as a handful of affordable hobby boards has grown into a diverse ecosystem powering automation,Armbian blogMichael RobinsonView the full article

If you have flashed working rkspi_loader.img to MTD before, i.e. from previous releases, you can keep using it with vendor kernel w/o updating MTD. The 26.2.1 image with vendor kernel uses new FIT u-boot loader as well, I doubt it will work with NVMe. BTW, I need 6.18 kernel and new u-boot and noticed that 26.2.1 image default to vendor kernel instead of 6.18 kernel now.

Hi all, I've spend a few evenings playing with the Armbian build-system and OPI4a, mainly trying to get Ethernet working reliably. There are two issues I've found: (almost) every second boot there are errors with PHY access (as if there was a race somewhere), neither DHCP nor static IPs are working (in scnerios when there are no issues with PHY). I've mainly focused on comparing Devicetrees to OrangePis' sources, and from newbie perspective, at least Ethernet part seems pretty complete. The next logical step would be to focus on the driver, but I figured, I'd ask if maybe someone has also been trying to work it out, and proceeded further? I'd appreciate any ideas of what I should check next.

Thanks to whom it may concern. I downloaded the today's version and it did connect..

Do you have rocket kernel with mesa that enable NPU working now? Or with rknn kernel? I'm a hobbyist working on a script to detect the NPU environment for ARM64 edge devices. Since I only have limited hardware, I’m looking for volunteers to run a quick test and see what is the environment on different NPU-equipped boards. This is a personal experimental project with no guarantees, just a survey to see what works and what doesn't. GitHub Repository: [npu-toolbox](https://github.com/flappyjet/npu-toolbox/) Test Script: `curl -sSL https://github.com/flappyjet/npu-toolbox/raw/refs/heads/main/scripts/npu_probe.sh | bash` Report Results: Please drop a comment in this survey issue: [Call for Testing: NPU detection Script](https://github.com/flappyjet/npu-toolbox/issues/1) Here's my A311d device info # npu-toolbox benchmark

Yeah i wanted to switch to noble minimal but it did not boot at all. noble i installed using armbian imager + online install. I think the problem is there are 3gb versions of opi 4 lts also and maybe all images are designed for those 3gb versions for compatability? Maybe there is some way to manually change that ram size config? inside device tree i could not find memory reg. Maybe it is related to uboot. Just adding i tested with noble desktop and it booted but still shows 3gb

This is quite old. Isn't there anything more recent? Perhaps this was fixed at some point already.

Hi! I have got Orange Pi 4 lts with 4 gb ram version. I made flashed sd card using Armbian-Imager using jammy_current_6.6.31 minimal. My problem is OS detects only 3gb (2.91) of 4gb. And i investigated and googled, spent more enough time to fix it but still fail. It is not likely reservation. also tried with bookworm the result is same 3gb. with noble minimal the board doesn't boot at all. what can i do to make os detect all 4 gbs. Maybe there was some similar issue before and i missed? Many thanks, appreciate any help!

Armbian 25.11.2 Noble XFCE (BSP Kernel: 6.1.115) + PanVk - mesa 26.0 (https://launchpad.net/~ernstp/+archive/ubuntu/mesaaco) + Box64 arm64 v0.4.1 ed9827920 (https://ryanfortner.github.io/box64-debs/) + wine-11.3-staging-tkg-amd64-wow64 (https://github.com/Kron4ek/Wine-Builds/releases/tag/11.3) + vkd3d-1.19 (wine's built-in) ~60fps@1024x576 (medium settings) Hades 2

Hi, I redownloaded the key and replaced it, and it works!

@technik007_cz your issue looks a little different from @Luna the Deer's. But have you both tried this?

Same issue there apt update Hit:1 http://deb.debian.org/debian trixie InRelease Hit:2 http://security.debian.org trixie-security InRelease Hit:3 http://deb.debian.org/debian trixie-updates InRelease Hit:4 http://deb.debian.org/debian trixie-backports InRelease Get:5 https://mirror.vinehost.net/armbian/apt trixie InRelease [39.4 kB] Err:5 https://mirror.vinehost.net/armbian/apt trixie InRelease Sub-process /usr/bin/sqv returned an error code (1), error message is: Signing key on DF00FAF1C577104B50BF1D0093D6889F9F0E78D5 is not bound: No binding signature at time 2026-02-26T23:23:28Z because: Policy rejected non-revocation signature (PositiveCertification) requiring second pre-image resistance because: SHA1 is not considered secure since 2026-02-01T00:00:00Z Missing key 8CFA83D13EB2181EEF5843E41EB30FAF236099FE, which is needed to verify signature. Fetched 39.4 kB in 4s (10.2 kB/s) All packages are up to date. Warning: An error occurred during the signature verification. The repository is not updated and the previous index files will be used. OpenPGP signature verification failed: https://mirror.vinehost.net/armbian/apt trixie InRelease: Sub-process /usr/bin/sqv returned an error code (1), error message is: Signing key on DF00FAF1C577104B50BF1D0093D6889F9F0E78D5 is not bound: No binding signature at time 2026-02-26T23:23:28Z because: Policy rejected non-revocation signature (PositiveCertification) requiring second pre-image resistance because: SHA1 is not considered secure since 2026-02-01T00:00:00Z Missing key 8CFA83D13EB2181EEF5843E41EB30FAF236099FE, which is needed to verify signature. Warning: Failed to fetch https://apt.armbian.com/dists/trixie/InRelease Sub-process /usr/bin/sqv returned an error code (1), error message is: Signing key on DF00FAF1C577104B50BF1D0093D6889F9F0E78D5 is not bound: No binding signature at time 2026-02-26T23:23:28Z because: Policy rejected non-revocation signature (PositiveCertification) requiring second pre-image resistance because: SHA1 is not considered secure since 2026-02-01T00:00:00Z Missing key 8CFA83D13EB2181EEF5843E41EB30FAF236099FE, which is needed to verify signature. Warning: Some index files failed to download. They have been ignored, or old ones used instead.

GPL requires source code be offered for any software distributed. Changes to the build system probably aren't covered, but a modified DTB probably is. The final image also has the Linux kernel and other GPL components, so that definitely is. Bottom line is that GPL is supposed to be permissive. DTS is easily gotten from DTB. And you're probably good as long as you're not being a jerk about it. If that bothers you, you could also make a overlay (dtbo) and just patch the DTB on boot. You'd avoid the custom image and also make it more clear that your DTB is tweaked.

While attempting, using AI, to see if there were DTS changes that could help this (spoiler: there aren't), it came up with some DTB changes that it claims need to be made for the board's stability. Maybe someone who understands these things can see if they got it right? Also, does the new allwinner update for armbian help/fix any of the cubie's USB/Power issues? https://github.com/armbian/build/pull/9381 a5e-armbian-DTS.patch

I finally managed to boot in Maskrom mode by shorting the clock pin (red circle on the picture) to ground, then loading the loader (rk3528_spl_loader_v1.07.104.bin) and u-boot with rkdeveloptool.

Dear @Werner , I have just created an issue on Github to keep track of this discussion: https://github.com/armbian/build/issues/9446 Is it safe to try possibly unsupported images on the NanoPC-T6 Plus ? It's a relatively expensive piece of kit and I'd rather not take any chances (new LPDDR5 RAM: possibly new memory timings, etc.) Best regards, Samuel

... problem is elsewhere. This is running: sudo dpkg-reconfigure tzdata in CLI (Armbian Noble), choosing region Atlantic, then:

bedna, see - i tried your way on a raspi 3B+ with the last minimal armbian and it worked like your case like a charm - on my Win11 with putty. If you look a bit back to my post with the attached "ssh_debian.txt" - that is what I get when I use a debian as ssh client! ssh client and the odroid ssh server cannot agree on how to connect UNLESS I dumb it down as seen in the command line - and then "1234" is not accepted.. So - armbian "odroid XU4" does not behave like "raspi"!

So I need to know what I have to do to get MIDI messages into my music app that I'm working on. I developed a cli synth app on Manjaro using SDL3 and Jack. Jack I used to connect to MIDI aka 'MIDI-Bridge' or MIDI Through and an external program to produce MIDI messages sent to MIDI Through. I am currently trying to do the same on the actual device I wish to use which is the OrangePi 4 LTS I have spare with just bare bones Armbian Server, no desktop. I have audio working through Pipewire. I am used to Jack just being available because Pipewire intercepts and handles the calls behind the scenes. I noticed that on Armbian I think I have to use the 'pw-jack' program or jack-play goes into an eternal and uninterruptible error loop (I have to kill it through another terminal). Note: I am developing this program in Zig. Am I doing all of this the wrong way? Is there some system process I have to start? Or something I haven't installed?

On the other hand, openSUSE Leap 16, KDE, it is not impossible:

Armbian config is just a wrapper to system tools that are dealing with this. We only change hardware related packages, the rest is stock Debian / Ubuntu. Is this problem solved somehow in Ubuntu? Probably the core of issue is originating from here: https://www.iana.org/ "Time zone of Iceland in Africa?" https://lists.iana.org/hyperkitty/list/tz@iana.org/message/QKQ27H23CTLFGJAQXJ24IVLRLQXHRLUA/ Interesting problem ...

Thank you for the resources, and you're right, I should probably be asking a lawyer. That said, the thing that's been throwing me off is that the GPL FAQ specifically states that and uses GCC as an example. Using GCC to build a binary does not make the binary subject to GPL. It seems like there's a similarity here to armbian/build which is a tool that produces an artifact and the code within customize-image.sh is my own code. This question really came up for me when I was modifying a DTS file to disable a peripheral on my device and was investigating whether that change would need to be open sourced. I found that the parent DTS file that I was patching was licensed under GPL 2.0 OR MIT, so if I'm reading everything correctly, it seems like neither the customize-image.sh script nor my DTS patches need to published and licensed under GPL 2.0. But I'm posing here in case I've egregiously misunderstood something. If there's nuance and subtlety here, where my assumptions above might be correct, or might not be depending on other factors, then I may need to get an attorney involved.

@Ryzer pcduino definitely works fine with dwc2. I cannot get OTG to work without it. I am not qulified to make any comments on whether this makes sense or is a coincidence, but if I don't load the module, it's doesn't work. For some reason, the mass storage gadget is still a little unstable in Windows (sometimes the File Explorer window shuts down for no reason), but on a Linux laptop, it is rock stable. It has also been running for two days in my car and doesn't seem to miss a beat there either (Linux host).