IgniteAMDGPU: GPU DevSecOps lab Provisioning Framework

A professional-grade, stateful provisioning system for building high-performance GPU-accelerated DevSecOps labs on Linux. Optimized for AMD hardware and the Polaris architecture.

🚀 Overview

This repository provides a modular, hardware-aware framework for bootstrapping a DevSecOps research environment. It automates the installation of the GPU compute stack (OpenCL/Vulkan), security toolkits, and AI inference engines while ensuring system stability and performance tracking.

💻 Compatibility Matrix

Supported Hardware (AMD GPU)

This framework is optimized for the AMD Polaris family but supports modern RDNA architectures:

• Polaris Series (10/20/30): Radeon RX 460, 470, 480 | RX 550, 560, 570, 580, 590
• Vega Series: Radeon VII, Vega 56, Vega 64
• Navi / RDNA Series: RX 5000 Series (5700 XT, etc.) | RX 6000 Series (6800, 6900 XT, etc.)

Tested Workstation Series

• Professional Workstations: Standard PCIe-equipped systems (Dell Precision, HP Z, Lenovo ThinkStation)
• Standard Desktops: OptiPlex, HP Elite, etc. (High-wattage PSU recommended for RX series)
• Generic: Any x86_64 system with an available PCIe x16 slot.

Supported Operating Systems

• Debian-based Distros: Ubuntu, Linux Mint, Parrot OS, Kali (Tested on Noble/Wilma)
• Ubuntu: 22.04 LTS, 24.04 LTS
• Debian: 11 (Bullseye), 12 (Bookworm)
• Parrot OS / Kali Linux: (Debian-based versions)

🛠️ Choose Your Tool

1. Lab Bootstrap v2 (gpu_lab_v2.sh)

Ideal for: Individual Researchers & Homelabs.

• Focus: Speed, simplicity, and performance tracking.
• Key Features: Hashcat benchmarking, attack profiles, stateful tracking, automated GRUB/Thermal tuning, and a Daily-Driver Safe Mode.

2. Enterprise Orchestrator v1 (provision_gpu_stack.sh)

Ideal for: Infrastructure & Production Labs.

• Focus: Safety, auditing, and structured recovery.
• Key Features: Tiered rollback registry, structured JSON logging, and decoupled functional validation.

---

🚦 Quick Start

# Clone the repository
git clone https://github.com/itxDeeni/IgniteAMDGPU.git
cd IgniteAMDGPU

# Run the Lab Bootstrap (v2)
chmod +x gpu_lab_v2.sh

# Recommended for Daily Drivers (No intrusive system changes)
./gpu_lab_v2.sh --safe-mode --profile=all

# Full Provisioning (Recommended for dedicated lab machines)
./gpu_lab_v2.sh --profile=all

🛡️ Reliability & Safety Features

• Idempotency: Re-running scripts is safe; they detect existing state (Apt, Gems, Git, Ollama) and skip completed tasks.
• Safety Modes: Includes a --safe-mode flag specifically for Daily Workstations to avoid intrusive kernel or security changes.
• Hardware-to-Driver Link: Real-time verification of amdgpu binding in sysfs.
• Reboot Detection: Smart flags prevent benchmarking on "dirty" driver states.
• Automated Tuning: Optional injection of power management parameters (ppfeaturemask) to unlock undervolting/fan control in CoreCtrl.