A CPU burn testing tool with real-time hardware monitoring for Linux systems.

• CPU Burn Testing: Spawns configurable worker goroutines performing intensive floating-point operations
• Hardware Monitoring: Real-time CPU frequency, temperature, and fan speed tracking
• Two Display Modes:
  • Line Mode: Simple text output with per-second statistics
  • Graph Mode: Interactive TUI with live graphs and dynamic worker control

go build -o goburn

# Run for 1 minute (line mode)
./goburn -duration=1m

# Run for 2 minutes with interactive TUI
./goburn -duration=2m -graph

• -duration: Test duration (default: 50s)
• -graph: Enable interactive TUI with graphs (default: false)

• + or =: Increase worker count
• - or _: Decrease worker count
• q or Ctrl+C: Quit

goburn/
├── main.go              # Entry point and CLI
├── hardware/
│   └── stats.go         # Hardware monitoring via Linux sysfs
├── worker/
│   └── pool.go          # Dynamic worker pool management
├── ui/
│   ├── line.go          # Simple line-based output
│   └── tui.go           # Interactive TUI with graphs
├── go.mod
└── README.md

Responsible for reading system hardware metrics from Linux sysfs:

• CPU frequency from /sys/devices/system/cpu/cpu*/cpufreq/
• Temperature from /sys/class/thermal/ and /sys/class/hwmon/
• Fan speeds from /sys/class/hwmon/*/fan*_input

Key Functions:

• Get(): Returns current hardware statistics
• Thread-safe and efficient file reading

Manages a dynamic pool of CPU-intensive worker goroutines:

• Workers perform floating-point math operations (math.Pow)
• Can dynamically add/remove workers at runtime
• Updates runtime.GOMAXPROCS() to match worker count
• Uses channels for graceful worker shutdown

Key Types:

• Pool: Manages worker lifecycle and shared counter

Key Methods:

• New(counter, count): Create pool with initial workers
• SetWorkers(n): Dynamically adjust worker count
• GetActiveCount(): Get current worker count

Provides two user interface implementations:

• Prints one line per second with current metrics
• Format: [elapsed] ops=XM/s | cpu=X/YMHz (Z%) | temp=X.XC | fans=X,YRPM
• Non-interactive, suitable for logging

• Full-screen TUI using Bubble Tea
• 2x2 grid of real-time graphs:
  • Operations per second
  • CPU frequency percentage
  • CPU temperature
  • Average fan speed
• Graphs automatically scale to terminal size
• Y-axis starts at 0, scales to theoretical maximum
• 60-second rolling history window

Key Features:

• Responsive layout adapts to terminal resize
• Color-coded output using Lip Gloss
• ASCII graphs using asciigraph

1. Add field to hardware.Stats struct in hardware/stats.go
2. Implement getter function (follow existing patterns)
3. Call getter in hardware.Get()
4. Update ui/line.go to format new metric
5. Add graph in ui/tui.go if desired

1. Create new file in ui/ package
2. Implement Run*Mode(counter, duration, startTime, ...) function
3. Add flag in main.go
4. Call from main() based on flag

Edit worker/pool.go:

• runWorker(): Change CPU load algorithm
• SetWorkers(): Modify scaling behavior
• Add new methods for worker control

• bubbletea: TUI framework
• lipgloss: Terminal styling
• asciigraph: ASCII graphs

Currently supports Linux only (requires sysfs for hardware monitoring).

• Each worker performs ~1 million operations before updating shared counter
• Atomic operations used for thread-safe counter updates
• Hardware stats read every second (I/O throttled)
• TUI updates at 1 Hz for smooth operation without excessive CPU usage

[Add your license here]