open source

How to build Ollama to run LLMs on RISC-V Linux

RISC-V is the new entrant into the SBC/low-end desktop space, and as I'm in possession of a HiFive Premier P550 motherboard, I am running it through my usual gauntlet of benchmarks—partly to see how fast it is, and partly to gauge how far along RISC-V support is in general across a wide swath of Linux software.

From my first tests on the VisionFive 2 back in 2023 to today, RISC-V has seen quite a bit of growth, fueled by economics, geopolitical wrangling, and developer interest.

The P550 uses the ESWIN EIC7700X SoC, and while it doesn't have a fast CPU, by modern standards, it is fast enough—and the system has enough RAM and IO—to run most modern Linux-y things. Including llama.cpp and Ollama!

Compiling Ollama for RISC-V Linux

I'm running Ubuntu 24.04.1 on my P550 board, and when I try running Ollama's simple install script, I get:

Raspberry Pi 500 uses QMK Firmware for built-in keyboard

I mentioned in my Pi 500 review Raspberry Pi is dogfooding their own microcontroller in the new Pi 500. An RP2040 sits next to the keyboard ribbon cable connector, and interfaces it through a USB port directly into the RP1 chip:

Raspberry Pi 500 PCB with RP2040 for keyboard input

In good news for keyboarding enthusiasts, the RP2040 seems to be flashed with the open-source QMK ('Quantum Mechanical Keyboard') Firmware. Thanks to a reader, 'M', who figured that out!

System76 Launch keyboard with RP2040 inside

Home Assistant Yellow - instant 2x IoT speedup with CM5

In a win for modular, private, local IoT, I just upgraded my Home Assistant Yellow from a Raspberry Pi Compute Module 4 to a Compute Module 5 this morning, and got an instant 2x speed boost.

Home Assistant Yellow upgraded to Pi CM5

I first posted about the Yellow in 2022, and walked through my smart-but-private HA Yellow setup in my Studio in a video last year.

Because I was running an eMMC CM4 in the Yellow before, I ran a full backup (and downloaded it), yanked the CM4, flashed HAOS to a new NVMe SSD, and plugged that and the CM5 into my Yellow. After running a Restore (it's a handy option right on the first page that appears when you access homeassistant.local), I was up and running like there was no difference at all—just everything was a little more snappy.

LLMs accelerated with eGPU on a Raspberry Pi 5

After a long journey getting AMD graphics cards working on the Raspberry Pi 5, we finally have a stable patch for the amdgpu Linux kernel driver, and it works on AMD RX 400, 500, 6000, and (current-generation) 7000-series GPUs.

With that, we also have stable Vulkan graphics and compute API support.

When I wrote about getting a Radeon Pro W7700 running on the Pi, I also mentioned AMD is not planning on supporting Arm with their ROCm GPU acceleration framework. At least not anytime soon.

Luckily, the Vulkan SDK can be used in its place, and in some cases even outperforms ROCm—especially on consumer cards where ROCm isn't even supported on x86!

Popular Rockchip SBC distro in limbo after maintainer burns out

Recently Joshua Riek posted he's dropping off from GitHub. If you haven't heard of him, he's one of the few reasons working with Linux on Rockchip SBCs is so much easier today than it was just a few years ago.

His Ubuntu Rockchip distribution is built for Ubuntu 22 and 24, and they've been maybe the most popular and stable way to run Ubuntu on Rockchip devices.

So popular, in fact, that manufacturers who use Rockchip, like Turing Pi, build their own official images on top of Joshua's.

XKCD Dependencies

Now, if you're reminded of XKCD #2347, yeah, I am too.

Use an External GPU on Raspberry Pi 5 for 4K Gaming

After I saw Pineboards 4K Pi 5 external GPU gaming demo at Maker Faire Hanover, I decided it was time to set up my GPU test rig and see how the Pi OS amdgpu Linux kernel patch is going.

GLmark2 running on Pi 5 with AMD RX 460 external GPU

I tested it out on a livestream over the weekend, but I thought I'd document the current state of the patch, how to apply it, and what else is left to do to get full external GPU support on the Raspberry Pi.

I also have a full video up with more demonstrations of the GPU in use, you can watch it below:

Sipeed NanoKVM: A RISC-V stick-on

Sipeed NanoKVM

This is the Sipeed NanoKVM. You stick it on your computer, plug in HDMI, USB, and the power button, and you get full remote control over the network—even if your computer locks up.

How did Sipeed make it so small, and so cheap? The 'full' kit above is about $50, while the cheapest competitors running PiKVM are closer to $200 and up!

This blog post is a lightly-edited transcript of the following video on my YouTube channel:

Saying a lot while saying nothing at all about Ansible AWX

A few days ago, the post Upcoming Changes to the AWX Project came across my feed. An innocuous title, but sometimes community-impacting changes are buried in posts like this. So, as an interested Ansible user, I read through the post.

In 1,610 words, almost nothing of substance was written.

A lot about how it's not 2014 anymore, so 2014-era architecture doesn't suit AWX. Then a big bold disclaimer at the bottom:

Before we conclude, we should be clear about what will not happen.

  • We are not changing the Ansible project
  • We are not adjusting our OSS license structure

Ultimately, we need to make some changes to the way our systems work and our projects are structured. Not a rewrite but a refactoring and restructuring of how some of the core components connect and communicate with each other.

Corporate Open Source is Dead

IBM is buying HashiCorp for $6.4 billion.

That's four months after HashiCorp rugpulled their entire development community and ditched open source for the 'Business Source License.'

As someone on Hacker News pointed out so eloquently:

IBM is like a juicer that takes all the delicious flavor out of a fruit

skywhopper replied:

HashiCorp has done a good job of pre-draining any flavor it once had.

Some people wonder if HashiCorp's decision to drop open source was because they wanted to juice the books for a higher price. I mean, six billion dollars? And they're not even a pointless AI company!