Skip to content

Installation

Requirements

  • Go 1.23+ (for building from source)
  • Redis server running on localhost:6379
  • GPUs with appropriate management tools:
  • NVIDIA GPUs: nvidia-smi available
  • AMD GPUs: amd-smi available (ROCm 5.7+)
  • System access to /proc filesystem or ps command for user detection

Dependencies

Go Dependencies

If building from source, Go dependencies are automatically managed:

go mod download

Redis Setup

Ubuntu/Debian

sudo apt update
sudo apt install redis-server
sudo systemctl start redis-server
sudo systemctl enable redis-server

macOS (via Homebrew)

brew install redis
brew services start redis

CentOS/RHEL/Fedora

sudo dnf install redis
sudo systemctl start redis
sudo systemctl enable redis

Verify Redis Installation

redis-cli ping
# Should return: PONG

GPU Drivers

NVIDIA GPUs

Ensure nvidia-smi is available:

nvidia-smi
# Should display GPU information

If not installed, install NVIDIA drivers for your system.

AMD GPUs

Ensure amd-smi is available:

amd-smi list
# Should display GPU information

If not installed, install ROCm drivers for your system:

Install canhazgpu

# Install directly from GitHub using Go
go install github.com/russellb/canhazgpu@latest

# The binary will be installed to $GOPATH/bin or $HOME/go/bin
# Make sure this directory is in your PATH
export PATH="$HOME/go/bin:$PATH"

Option 2: Build from Source

# Clone the repository
git clone https://github.com/russellb/canhazgpu.git
cd canhazgpu

# Build and install using Makefile
make install

# Optional: Install documentation dependencies for building docs
make docs-deps

Option 3: Pre-built Binary

# Download pre-built binary (when available)
wget https://github.com/russellb/canhazgpu/releases/latest/download/canhazgpu
chmod +x canhazgpu

# Download bash completion script (optional)
wget https://raw.githubusercontent.com/russellb/canhazgpu/main/autocomplete_canhazgpu.sh

# Install system-wide
sudo cp canhazgpu /usr/local/bin/
sudo cp autocomplete_canhazgpu.sh /etc/bash_completion.d/

# Optional: Create short alias symlink
sudo ln -s /usr/local/bin/canhazgpu /usr/local/bin/chg

Option 4: Local Installation

# Keep in local directory and add to PATH
export PATH="$PWD:$PATH"

Bash Completion

The bash completion script provides tab completion for canhazgpu commands and options. It also supports the short alias chg if you've created the symlink.

Required for canhazgpu run Commands

Installing bash completion is required for proper tab completion when using commands with canhazgpu run. Without it, bash completion won't work for the commands you run after the -- separator.

Enable Completion

After installing the completion script to /etc/bash_completion.d/, enable it:

# Reload bash completion
source /etc/bash_completion

# Or restart your shell
exec bash

Usage Examples

With bash completion enabled, you can use tab completion:

# Complete commands
canhazgpu <TAB>
# Shows: admin  release  report  reserve  run  status  web

# Complete commands (works with chg alias too)
chg <TAB>
# Shows: admin  release  report  reserve  run  status  web

# Complete options
canhazgpu run --<TAB>
# Shows: --gpus  --help

# Complete duration formats
canhazgpu reserve --duration <TAB>
# Shows common duration examples

# Complete commands after 'canhazgpu run --'
canhazgpu run --gpus 1 -- python <TAB>
# Shows available Python files and completion

# Complete GPU management tool options
canhazgpu run --gpus 1 -- nvidia-smi --<TAB>  # For NVIDIA
# Shows nvidia-smi options
canhazgpu run --gpus 1 -- amd-smi --<TAB>     # For AMD
# Shows amd-smi options

Manual Installation

If the automatic installation doesn't work, you can source the completion script manually:

# Add to your ~/.bashrc
echo "source /path/to/autocomplete_canhazgpu.sh" >> ~/.bashrc
source ~/.bashrc

Verification

Test the installation:

canhazgpu --help

You should see the help output with available commands.

Test bash completion (if installed): 

canhazgpu <TAB><TAB>

Should show available commands.

Next Steps