Robin Voetter

light-rocm-runtime. A light ROCm runtime

cmpunlocker. A tool to unlobotomize your NVIDIA card!

ninjatracing. Convert .ninja_log files to chrome's about:tracing format.

cutile-rs. cuTile Rust provides a safe, tile-based kernel programming DSL for the Rust programming language. It features a safe host-side API for passing tensors to asynchronously executed kernel functions.

tree-sitter-amdgpu-asm. A Tree-Sitter grammar to provide highlighting for amdgpu assembly

semble. Fast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+read

auto-identity-remove. Automated data broker opt-out runner — removes your personal info from 30+ people-search sites on a monthly schedule

Ashes. Drop-in replacement for Vulkan shared library, for older hardware compatibility

cuda-oxide. cuda-oxide is an experimental Rust-to-CUDA compiler that lets you write (SIMT) GPU kernels in safe(ish), idiomatic Rust. It compiles standard Rust code directly to PTX — no DSLs, no foreign language bindings, just Rust.

Snort-GPU. Snort GPU

midint-arithmetic-division. An implementation of division for midsized integers in CUDA

proth20. An OpenCL implementation of Proth's Theorem

CLVizulayer. OpenCL layer creating a DOT file of the dependency graph of asynchronous device tasks

Satty. Satty - Modern Screenshot Annotation.

arnis. Generate any location from the real world in Minecraft with a high level of detail.

zephyr. Primary Git Repository for the Zephyr Project. Zephyr is a new generation, scalable, optimized, secure RTOS for multiple hardware architectures.

tiny-gpu-compiler. An MLIR-based compiler that takes GPU kernels and compiles them to real hardware instructions. Interactive web visualizer included.

FlyDSL. FlyDSL is the Python front‑end of the project: a Flexible Layout Python DSL for expressing tiling, partitioning, data movement, and kernel structure at a high level.

Booth. Open-source CUDA, Triton and HIP compiler targeting multiple GPU and CPU architectures.

rocm-kpack. Prototype for a kernel packing library

ipmitool. An open-source tool for controlling IPMI-enabled systems

msvcup. Hermetic install of MSVC/SDK from the CLI

roc-optiq. A visualizer for the ROCm Profiler Tools

zerobrew. A 5-20x faster experimental Homebrew alternative

hipFile. [DEPRECATED] Moved to ROCm/rocm-systems repo

rheo. Typesetting and static site engine based on Typst

whosthere. Local Area Network discovery tool with an interactive Terminal User Interface (TUI) written in Go. Discover, explore, and understand your LAN in an intuitive way. Knock Knock.. who's there? 🚪

hide-youtube-shorts. Browser add-on that hides YouTube-shorts videos

templight. Templight is a Clang-based tool to profile the time and memory consumption of template instantiations and to perform interactive debugging sessions to gain introspection into the template instantiation process.

nix-config. meow

ClangBuildAnalyzer. Clang build analysis tool using -ftime-trace

marek-build. Marek's approach to building AMD GPU drivers for driver development

HipKittens. Fast and Furious AMD Kernels

nixos-raspberrypi. Nix flake for a fully declarative NixOS on Raspberry Pi

nohang. A sophisticated low memory handler for Linux

nvcc-llvm-ir. Enabling on-the-fly manipulations with LLVM IR code of CUDA sources

iongraph. TypeScript

LACT. Linux GPU Configuration And Monitoring Tool

ssh3. SSH3: faster and rich secure shell using HTTP/3, checkout our article here: https://arxiv.org/abs/2312.08396 and our Internet-Draft: https://datatracker.ietf.org/doc/draft-michel-ssh3/

Asus-ROG-Aml-Deep-Dive. A deep dive into the ACPI.sys DPC latency problems on Asus ROG laptops

gpu-kill. Manage your GPUs across NVIDIA, AMD, Intel, and Apple Silicon systems.

warewulf-node-images. Example node images for use with Warewulf 4

warewulf-slurmd. Hardened Slurm compute node image for Warewulf 4 HPC clusters. Ubuntu 24.04 with optional kernel/NVIDIA support, CIS security hardening.

warewulf-extras. Supplemental material for my talk on using Warewulf at ASU

kerneldb. A library for interacting with CDNA code objects to support performance analysis

profilerpedia. Catalog of profilers, profile data formats, data format converters, and profile analysis UIs.

omniprobe. Simply log all kernel durations

rocr_debug_agent. The ROCdebug-agent is a library that can be loaded by ROCm Platform Runtime to provide some debugging functionality.

services-flake. NixOS-like services for Nix flakes [maintainer=@shivaraj-bh]

kernelbot. Write a fast kernel and see how you compare against the best humans and AI on gpumode.com

rocm-systems. super repo for rocm systems projects

reverie. An ergonomic and safe syscall interception framework for Linux.

warpgate. Fully transparent SSH, HTTPS, Kubernetes, MySQL and Postgres bastion/PAM that doesn't need additional client-side software

ECE298A-TPU. A custom AI chip to be taped out soon!

cppreference-doc. C++ standard library reference

rust-gpu-chimera. Demo project showing a single Rust codebase running on CPU and directly on GPUs

scrcpy. Display and control your Android device

gpuhammer. Cuda

pangolin. Identity-aware VPN and tunneled reverse proxy for remote access based on WireGuard®.

faiss. A library for efficient similarity search and clustering of dense vectors.

Graphite. Community-built comprehensive 2D content creation appplication for graphic design, digital art, and interactive real-time motion graphics powered by a node-based procedural graphics engine

zig-sdl3-gpu-examples. Experimental SDL3 GPU examples using zig shaders.

neutrino. C

aqlprofile. [DEPRECATED] Moved to ROCm/rocm-systems repo

rocprof-trace-decoder. C

xan. The CSV magician

RadeonFlow_Kernels. Efficient implementation of DeepSeek Ops (Blockwise FP8 GEMM, MoE, and MLA) for AMD Instinct MI300X

amd_challenge_solutions. C++

amd-challenge. Global #4 on FP8 GEMM on MI300X. https://akashkarnatak.github.io/amd-challenge

rocprof-compute-viewer. C++

rocm-rs. Rust

rocm-libraries. super repo for rocm libraries

TheRock. The HIP Environment and ROCm Kit - A lightweight open source build system for HIP and ROCm

libadrenotools. A library for applying rootless Adreno GPU driver modifications/replacements

firefox. The official repository of Mozilla's Firefox web browser.

vtree. Futhark implementation of vtrees - data parallel operations on trees based on Euler tours

cuda-side-boost. Cuda

rocmProfileData. Python

attune. Fast and secure Linux package publishing.

opkssh. opkssh (OpenPubkey SSH)

openpubkey. Reference implementation of OpenPubkey

amdgpu-rs. Examples and support libraries for the amdgpu Rust target

Slurm-web. Open source web interface for Slurm HPC & AI clusters

brutus. Modeling stellar photometry with "brute force" methods

quickreduce. QuickReduce is a performant all-reduce library designed for AMD ROCm that supports inline compression.

gpu-fryer. Where GPUs get cooked 👩‍🍳🔥

caligula. A user-friendly, lightweight TUI for disk imaging

DeepGEMM. DeepGEMM: clean and efficient BLAS kernel library on GPU

amd_matrix_instruction_calculator. A tool for generating information about the matrix multiplication instructions in AMD Radeon™ and AMD Instinct™ accelerators

Gaudi-tutorials. Tutorials for running models on First-gen Gaudi and Gaudi2 for Training and Inference. The source files for the tutorials on https://developer.habana.ai/

fp32_sgemm_amd. Super fast FP32 matrix multiplication on RDNA3

clr. Checker for Lifetimes and other Refinement types

timemory. C++

Bear. Generate compile_commands.json for any C or C++ build

AMG. Algebraic multigrid benchmark

uscope. μscope 🔬

tilelang. Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

CLRX-mirror. CLRadeonExtender (GCN assembler, Radeon assembler) mirror

isd. isd (interactive systemd) – a better way to work with systemd units

pwndbg. Exploit Development and Reverse Engineering with GDB & LLDB Made Easy

ghostty. 👻 Ghostty is a fast, feature-rich, and cross-platform terminal emulator that uses platform-native UI and GPU acceleration.

ml.nix. Nix

NVPTX-SPIRV-Translator. The translator that supports translating NVPTX to SPIR-V. This translator is modified from LLVM-SPIR-V Translator.

aoc24. I don't have time for this.

hipRAND. [DEPRECATED] Moved to ROCm/rocm-libraries repo

hipCUB. [DEPRECATED] Moved to ROCm/rocm-libraries repo

rocThrust. [DEPRECATED] Moved to ROCm/rocm-libraries repo

rocPRIM. [DEPRECATED] Moved to ROCm/rocm-libraries repo

rocRAND. [DEPRECATED] Moved to ROCm/rocm-libraries repo

buildkit-nix. Nix derivations as Dockerfiles (`docker build -f default.nix .`)

nix-du. Visualise which gc-roots to delete to free some space in your nix store

librespot. Open Source Spotify client library

gcnasm. amdgpu example code in hip/asm

V3DLib. C++ library for programming the VideoCore GPU on all Raspberry Pi's.

kernel_launcher. Using C++ magic to capture CUDA kernels and tune them with Kernel Tuner

z80-sans. OpenType font that disassembles Z80 instructions

vello. A GPU compute-centric 2D renderer.

mantis. Mantis is a unified infrastructure as code framework that replaces Terraform and Helm

bootloader-unlock-wall-of-shame. Keeping track of companies that "care about your data 🥺"

system-manager. Manage system config using nix on any distro

lupine. LUPINE is a GPU over IP bridge allowing GPUs on remote machines to be attached to CPU-only machines.

nixos-apple-silicon. Resources to install NixOS bare metal on Apple Silicon Macs [maintainers=@tpwrules,@flokli,@yuyuyureka]

doomgeneric. A GPU port of DOOM

running-nvidia-sxm-gpus-in-consumer-pcs. Running SXM2/SXM3/SXM4 NVidia data center GPUs in consumer PCs

nix-system-graphics. Run graphics accelerated programs built with Nix on any Linux distribution.

system-manager. Manage system config using nix on any distro

smol-v. SMOL-V: like Vulkan/Khronos SPIR-V, but smaller.

colmena. A simple, stateless NixOS deployment tool [maintainer=@zhaofengli, @stepbrobd, @NickCao]

zml. Any model. Any hardware. Zero compromise. Built with @ziglang / @openxla / MLIR / @bazelbuild

zig-vulkan-triangle. simple triangle displayed using vulkan, xcb, and zig

glslang-zig. Glslang ported to the Zig build system.

shader_compiler. Glslang ported to be usable via Zig's build system.

gitlab-ci-local. Tired of pushing to test your .gitlab-ci.yml?

fresh-bot. The opposite of Stale. A bot against stale bots.

sass_graph. NVIDIA Sass binary CFG graph visualization tool.

isa_spec_manager. Utilities for accessing AMD's Machine-Readable GPU ISA Specifications.

alien. ALIEN is a CUDA-powered artificial life simulation program.

Blendersito. Una version de Blender creada desde cero para Symbian

glsl_analyzer. Language server for GLSL (autocomplete, goto-definition, formatter, and more)

spice. Fine-grained parallelism with sub-nanosecond overhead in Zig

LibreCuda. C

gpuperftests. Derived from Nemes' gpuperftests

HVM2. A massively parallel, optimal functional runtime in Rust

NativeLinux. Linux in the Windows Native Mode (chkdsk screen env)??

cubecl. Multi-platform high-performance compute language extension for Rust.

typst. A markup-based typesetting system that is powerful and easy to learn.

openxr-zig. OpenXR binding generator for Zig

jpeggpu. Low-latency CUDA JPEG decoder by parallelizing Huffman decoding

gpu.cpp. A lightweight library for portable low-level GPU computation using WebGPU.

nv_isa_solver. Nvidia Instruction Set Specification Generator

xpra-html5. HTML5 client for Xpra

hevi. ⚠️ MIGRATED TO CODEBERG ⚠️

astroz. Astrodynamics and Spacecraft Toolkit. Features fast orbit prop, celestial precession, CCSDS parsing, RF parsing, fits image parsing, and more!

composable_kernel. [DEPRECATED] Moved to ROCm/rocm-libraries repo. NOTE: develop branch is maintained as a read-only mirror

dom3d. Browser extension to view and debug the DOM in 3D space.

proxmox-nixos. The Proxmox Hypervisor, on NixOS [maintainers=@camillemndn @julienmalka]

unicorn. Unicorn CPU emulator framework (ARM, AArch64, M68K, Mips, Sparc, PowerPC, RiscV, S390x, TriCore, X86)

setup-zig. Install a Zig compiler for usage in GitHub Actions workflows. Read-only mirror of https://codeberg.org/mlugg/setup-zig.

iglunix. Linux (and other kernels) distro with no GNU components

gpuocelot. GPUOcelot: A dynamic compilation framework for PTX

ladybird. Truly independent web browser

rocm_sdk_builder. Shell

terraform-provider-dominos. The Terraform plugin for the Dominos Pizza provider.

ThunderKittens. Tile primitives for speedy kernels

utpx. UTPX (Userspace Transparent Paging Extension) is a proof-of-concept LD_PRELOAD library that accelerates HIP managed allocations on systems without XNACK or with XNACK disabled.

mgmt. Next generation distributed, event-driven, parallel config management!

Ashet-OS. Retro-inspired operating system designed to be learnable and hackable by its users

Bend. A massively parallel, high-level programming language

tiny11builder. Scripts to build a trimmed-down Windows 11 image.

SnekCraft. Java

dvui. Immediate Zig GUI for Apps and Games

radeon-tools. Radeon reverse engineering tools

vendor-reset. Linux kernel vendor specific hardware reset module for sequences that are too complex/complicated to land in pci_quirks.c

7900xtx. Python

HIPRT. C++

wasm2spirv. Compile your WebAssembly programs into SPIR-V shaders

cbscript. CBScript for Minecraft

nixbsd. An unofficial NixOS fork with a FreeBSD kernel

calyx. Intermediate Language (IL) for Hardware Accelerator Generators

nixpkgs-terraform. A collection of Terraform versions that are automatically updated.

mach-gpu. mach/gpu: truly cross-platform WebGPU graphics for Zig

ZLUDA. CUDA on non-NVIDIA GPUs

mach-dxcompiler. DXC built using Zig

ReBarUEFI. Resizable BAR for (almost) any UEFI system

zpoly. zig polynomial tools and root finding

snek. Snek programming language for tiny systems

Nabla. Vulkan, OptiX and CUDA Interoperation Modular Rendering Library and Framework for PC/Linux/Android

exercise-timer. Timer app for high intensity interval training

SSH-Snake. SSH-Snake is a self-propagating, self-replicating, file-less script that automates the post-exploitation task of SSH private key and host discovery.

CuPBoP. A framework that support executing unmodified CUDA source code on non-NVIDIA devices.

SoilMachine. Advanced, modular, coupled geomorpohology simulator for real-time procedural terrain generation.

SimpleHydrology. Procedural Hydrology / River / Lake Simulation

zig-llm. Zig Llama2 implementation in WGSL with Mach backend.

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee. eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

sargantana. SystemVerilog

NixOS-DNS. Manage your dns zones with NixOS

Halide. a language for fast, portable data-parallel computation

stuxnet. Public open-source code of malware Stuxnet (aka MyRTUs).

zig-bounded-array. BoundedArray module for Zig.

commet. Your space to connect