RISC-V CPU on FPGA
an invention by mkoutny
Updated about 2 years ago. No love. 2 followers.

Project Description

Software engineers should be proficient in their area of expertise and also have overview of adjacent abstraction layers to understand the world in better context. Despite SUSE is not a hardware company, this project aims at implementing a simple RISC-V processor on a hobby FPGA board.

More details are in project repo's README.

Goal for this Hackweek

  • Get familiar with FPGA tools
  • Peek into HDL (hardware definition languages)
  • Design the CPU using SystemVerilog
  • Create a demo program to run on the CPU
  • Run the circuit in a simulator and eventually on the FPGA too
  • Implement some non-trivial features like pipelining or some instruction set extension

Resources

Looking for hackers with the skills:

risc-v fpga clang verilog

This project is part of:

Hack Week 23

Activity

  • about 2 years ago: PPavlu joined this project.
  • about 2 years ago: mkoutny added keyword "risc-v" to this project.
  • about 2 years ago: mkoutny added keyword "fpga" to this project.
  • about 2 years ago: mkoutny added keyword "clang" to this project.
  • about 2 years ago: mkoutny added keyword "verilog" to this project.
  • about 2 years ago: mkoutny started this project.
  • about 2 years ago: mkoutny originated this project.


    • Comments

    • mkoutny
      about 2 years ago by mkoutny | Reply

      We implemented several blocks of the circuitry:

      • control (roughly fetch/decode),
      • register file,
      • ALU,
      • block RAM,
      • memory control (for unaligned access).

      We also implemented memory mapped IO (using serial over USB block provided by TinyFPGA's library).

      Additionally, we made a simple demo program and tooling to cross-compile it on non-RISC-V machine and prepare it to "load" it to FPGA board's memory.

      Every circuit block is accompanied with (simulator) testsuite and they pass \o/ We didn't manage to run the circuit on FPGA, probably because of memory data preloading or cumbersome IO. That will need more debugging in the future. (Hence we also didn't get down to any of the non-trivial features.)

    Similar Projects

    pudc - A PID 1 process that barks to the internet by mssola

    Description

    As a fun exercise in order to dig deeper into the Linux kernel, its interfaces, the RISC-V architecture, and all the dragons in between; I'm building a blog site cooked like this:

    • The backend is written in a mixture of C and RISC-V assembly.
    • The backend is actually PID1 (for real, not within a container).
    • We poll and parse incoming HTTP requests ourselves.
    • The frontend is a mere HTML page with htmx.

    The project is meant to be Linux-specific, so I'm going to use io_uring, pidfs, namespaces, and Linux-specific features in order to drive all of this.

    I'm open for suggestions and so on, but this is meant to be a solo project, as this is more of a learning exercise for me than anything else.

    Goals

    • Have a better understanding of different Linux features from user space down to the kernel internals.
    • Most importantly: have fun.

    Resources

    Create openSUSE images for Arm and RISC-V boards by avicenzi

    Project Description

    Create openSUSE images (or test generic EFI images) for Arm and RISC-V boards that are not yet supported.

    Goal for Hackweek

    Create bootable images of Tumbleweed for SBCs that currently have no images available or are untested.

    Consider generic EFI images where possible, as some boards can hold a bootloader.

    Document in the openSUSE Wiki how to flash and use the image for a given board.

    Hack Week 25

    Hack Week 24

    Hack Week 23

    Hack Week 22

    Hack Week 21

    Resources

    Tumbleweed on Mars-CM (RISC-V board) by ph03nix

    RISC-V is awesome, Tumbleweed is awesome, chocolate cake is awesome. I'm planning to combine all of them in one project.

    Project Description

    I recently purchased a MILK-V Mars CM and managed to setup it up already using Debian Linux. My project for this Hackweek is to see how far I can get to run Tumbleweed on this compute module board.

    Goal for this Hackweek

    • Run Tumbleweed on the Compute Module

    Resources

    • http://milkv.io/mars-cm
    • https://en.opensuse.org/HCL:VisionFive2

    Reassess HiFive Premier P550 board (for RISC-V virtualization) by a_faerber

    Description

    With growing interest in the RISC-V instruction set architecture, we need to re-evaluate ways of building packages for it:

    Currently openSUSE OBS is using x86_64 build workers, using QEMU userspace-level (syscall) emulation inside KVM VMs. Occasionally this setup causes build failures, due to timing differences or incomplete emulation. Andreas Schwab and others have collected workarounds in projects like openSUSE:Factory:RISCV to deal with some of those issues.

    Ideally we would be using native riscv64 KVM VMs instead. This requires CPUs with the H extension. Two generally available development boards feature the ESWIN 7700X System-on-Chip with SiFive P550 CPUs, HiFive Premier P550 and Milk-V Megrez. We've had access to the HiFive Premier P550 for some time now, but the early version (based on Yocto) had issues with the bootloader, and reportedly later boards were booting to a dracut emergency shell for lack of block device drivers.

    Goals

    • Update the boot firmware
    • Test whether and how far openSUSE Tumbleweed boots

    Results

    • Boot firmware image 2025.11.00 successfully flashed onto board
      • Enables UEFI boot in U-Boot by default
      • U-Boot's embedded Flat Device Tree is lacking a timebase-frequency, required for recent (6.16.3) mainline kernels (panic leading to reset, visible via earlycon=sbi)
    • Tested eswin/eic7700-hifive-premier-p550.dtb from Ubuntu 2025.11.00 image
      • Allows to boot past the above panic, but times out in JeOS image while waiting for block device, dropping to dracut emergency shell
      • No devices shown in lsblk -- 6.16 appears to be lacking device drivers still

    Resources