compile

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:

Ansible runs on Python, and Python runs on... well pretty much everything. Including newer RISC-V machines.

But Ansible has a lot of dependencies, and some of these dependencies have caused frustration from time to time on x86 and Arm (so having issues with a dependency is just a way of life when you enter dependency hell)... but in this case, for the past few months, I've never had luck installing Ansible from PyPI (Python's Package Index) on any RISC-V system, using pip install ansible.

I prefer installing this way (rather than compiling from source or from system packages) because it generally gets the latest version of Ansible, with an easy upgrade/downgrade path. It's also easy to add ansible to a Python requirements.txt file and install it alongside other package dependencies.

UPDATE 2025-02: It seems that at least under Ubuntu's latest releases, Ansible installs with a simple pip install ansible again, no need for the extra steps below. Yay!

Raspberry Pi OS isn't really built to be a server OS; the main goals are stability and support for educational content. But that doesn't mean people like me don't use and abuse it to do just about anything.

In my case, I've been doing a lot of network testing lately—first with an Intel I340-T4 PCIe interface for 4.15 Gbps of networking, and more recently (yesterday, in fact!) with a Rosewill 2.5 GbE PCIe NIC.

And since the Pi's BCM2711 SoC is somewhat limited, it can't seem to pump through many Gbps of bandwidth without hitting IRQ limits, and queueing up packets.

In the case of the 2.5G NIC, I was seeing it max out around 1.92 Gpbs, and I just wouldn't accept that (at least not for a raw benchmark). Running atop, I noticed that during testing, the IRQ interrupts would max out at 99% on one CPU core—and it seems like it may be impossible to distribute interrupts across all four cores on the BCM2711.

After doing a video testing different external GPUs on a Raspberry Pi last week, I realized two things:

1. Compiling the Linux kernel on a Raspberry Pi is slow. It took 54 minutes, and I ended up doing it 7 times during the course of testing for that video.
2. If you ever want to figure out a better way to do something, write a blog post or create a video showing the less optimal way of doing it.

To the second point, about every fifth comment was telling me to cross-compile Linux on a faster machine instead of doing it on the Pi itself. For example:

And on the Pi Forums, it seems like nobody worth their salt compiles the kernel on the Pi either, so I figured—since I'm probably going to have to do it again another thousand times in my life—I might as well put together a guide for how to do it on a Mac.