🛠️ Linux Driver Evaluation System

This project implements an automated evaluation pipeline for Linux character device driver code — including code generated by LLMs.
The system scores each driver file on:

• ✅ Compilation success
• ✅ Style compliance
• ✅ Security heuristics

It outputs a detailed JSON report with component-wise scores.

---

📌 Features

• ⚙️ Compilation check using gcc
• 🎯 Style compliance via checkpatch.pl (Linux kernel style)
• 🔐 Security scan for:
  • Buffer overflows
  • Race condition indicators
  • Input validation issues
• 🤖 LLM-generated driver evaluation
• 🧮 Score calculation based on weighted metrics
• 🧪 Bonus: Runtime test simulation
• 💡 LLM fine-tuning suggestions

---

🚀 How to Use

1. Upload your .c driver file in Colab (e.g., char_driver_sample.c)

2. Run each cell in h2loop.ipynb:

   • compile_driver()
   • check_style()
   • run_security_checks()
   • calculate_score()

3. Final output:
   📄 evaluation_result.json
   Includes compilation, style, and security scores with total

---

🧪 Test Files Included

• char_driver_sample.c — base test driver
• llm_generated_driver.c — ChatGPT-generated driver
• test_driver_sim.c — simulates runtime test

---

📊 Sample Output

{
  "compilation": {"success": false, "warnings": 0, "errors": 1},
  "code_quality": {"style_score": 1.0},
  "security": {
    "buffer_safety": 1.0,
    "race_conditions": 0.5,
    "input_validation": 0.4
  },
  "overall_score": 35.83
}

---

🧠 Bonus Tasks

1. Simulated Runtime Test

A user-space simulation (test_driver_sim.c) mimics interaction with the driver (read/write buffer). ➡️ Output saved in results/runtime_test_output.txt

2. LLM Evaluation Suggestions

See: LLM_Suggestions file for notes on fine-tuning, prompt engineering, and evaluation loop design.

---

📚 Documentation

• ARCHITECTURE
• RUBRICS
• USER_GUIDE
• LLM_SUGGESTIONS

---

📁 Repo Structure

h2loop.ipynb                 → Colab-based evaluation pipeline
char_driver_sample.c        → Sample driver file
llm_generated_driver.c      → LLM-generated test driver
test_driver_sim.c           → Runtime simulation test
evaluation_result.json      → Sample output JSON
ARCHITECTURE / RUBRICS / USER_GUIDE / LLM_Suggestions → Documentation files

---

🎥 Demo Recording

👉 Watch full walkthrough on Loom

---

👤 Author

Divyansh Gautam Automated evaluation pipeline for Linux device-driver code — including LLM-generated drivers.