Home

Building GPP

The following pages give information on GPP's build system and dependencies

• GPP Quick Start Guide

  1. Installation
  2. Internal Testing
  3. Kernel Autotuning
  4. Debugging
  5. Manually Running Basic Tests

• GPP Build With CMake

  1. GPP and CMake
  2. Obtaining CMake
  3. Building GPP using CMake
  4. Reducing GPP build time
  5. Building GPP with clang-cuda
  6. Building GPP with HIP/ROCm

• GPP Environment Variables

Multi-GPU GPP

These pages give information on how to build GPP using MPI, including some machine-specific instructions

• Multi GPU Support
  1. Compiling for Multi-GPU
  2. Running
  3. Running GPP's Tests
  4. Multi-GPU Emulation
  5. Peer-to-peer Communication
  6. GPU Direct RDMA and CUDA-aware MPI
  7. Maximizing GDR Performance
  8. Dependence on CUDA_DEVICE_MAX_CONNECTIONS
  9. Low-level Details
  10. Dslash Policy Tuning
  11. Dslash Component Benchmarking
• Installing MVAPICH2 GDR
• Split Grid

Using GPP In Your applications

• GPP Interface
• GPP Parameter Structures
• GPP Data Structures And Types
• Best Practices

GPP Development

Contributing to GPP

• Updating the GPP Interface
• Adding new GPP features
• Writing GPP Kernels
• Coding Conventions and Style

Resources

• GPP Development model Some Rules and Best Practices for writing code and use of git
• Continuous Integration
• GPP Release Procedure
• GPP Release Testing
• GPP Debugging
• Timing GPP build
• External GPP mods

Case Studies: Pedagogical Instructions On Using Some of GPP's Key Features

Below we have included some important case studies, designed to be guides on how to get the most out of GPP solvers for different use cases in modern LQCD.

• Tensor Cores