itzmeanjan

ML-DSA was previously known as Dilithium (https://pq-crystals.org/dilithium/index.shtml) and it has been standardized by NIST. The NIST FIPS 204 standard lives @ https://doi.org/10.6028/NIST.FIPS.204.

Note

These KATs are used to test functional correctness & conformance of ML-DSA implementation @ https://github.com/itzmeanjan/ml-dsa.

Using ML-DSA Known Answer Tests

Simply clone this gist repository

itzmeanjan

[package]
name = "gpu-matrix-multiplication"
version = "0.1.0"
edition = "2024"
rust-version = "1.85.0"

[dependencies]
vulkano = "=0.35.1"
vulkano-shaders = "=0.35.0"
rand = "=0.8.4"

itzmeanjan

#!/bin/bash

# System update
sudo apt-get update
sudo apt-get upgrade -y
sudo apt-get autoremove -y

# Install system development tools
sudo apt-get install build-essential -y
sudo apt-get install cmake -y

itzmeanjan

Background

I'm demonstrating following on a machine, running Ubuntu 22.04 with GNU/Linux kernel 5.19.0.

$ lsb_release -a
No LSB modules are available.
Distributor ID:	Ubuntu
Description:	Ubuntu 22.04.2 LTS
Release:	22.04

itzmeanjan

#!/usr/bin/bash

# Script version of "Getting Started with the Linux Tarball Vulkan SDK" guide @ https://vulkan.lunarg.com/doc/sdk/1.4.335.0/linux/getting_started.html.
# Targets Ubuntu, tested to be working on Ubuntu 25.10.
# Source of truth for Vulkan SDK version and checksum @ https://vulkan.lunarg.com/sdk/home#linux.

EXPECTED_SHA256_CHECKSUM="ccab8047f33ef848e3928c7fdd19987b35da3085ad20178f473acc230fa3c5f2"
VULKAN_SDK_VERSION="1.4.335.0"
VULKAN_SDK_TARBALL_DOWNLOAD_URL="https://sdk.lunarg.com/sdk/download/${VULKAN_SDK_VERSION}/linux/vulkansdk-linux-x86_64-${VULKAN_SDK_VERSION}.tar.xz"
VULKAN_SDK_TARBALL=$(realpath vulkansdk-linux-x86_64-${VULKAN_SDK_VERSION}.tar.xz)

itzmeanjan

1. Clone this gist repository.
2. Setup python virtual environment.

# `cd` into this gist
python3 -m venv .
source bin/activate

3. Fetch dependencies.

python -m pip install -r requirements.txt

itzmeanjan

blake3==1.0.5
pillow==11.3.0

itzmeanjan

#include <CL/sycl.hpp>
#include <chrono>
#include <complex>
#include <iostream>

using namespace sycl;

typedef std::complex<double> cmplx;

constexpr uint Y = 1 << 11;

itzmeanjan

How to ?

1. Download ENS labels JSON database from https://github.com/adraffy/ens-labels.git (retrieved at commit d1d85346).
2. Download this gist and unzip. Then move the downloaded labels.json file inside the unzipped project directory.
3. Inside unzipped project directory, create a Python virtual environment.

python -m venv .
source bin/activate

4. Download dependencies using pip.

itzmeanjan

#!/usr/bin/python3

from math import ceil
from typing import List, Tuple
from random import randint


def bit_count(num: int) -> int:
    '''
    Same as len(bin(num)[2:])