|
Download CONTRIBUTING.md from elyanlabs/llama-cpp-power8: direct link, hf CLI and curl.
- Browser
- Download file 1.58 kB
-
https://huggingface.co/elyanlabs/llama-cpp-power8/resolve/main/CONTRIBUTING.md
- Command line
-
hf download hf://elyanlabs/llama-cpp-power8/CONTRIBUTING.md
-
curl -L -o CONTRIBUTING.md https://huggingface.co/elyanlabs/llama-cpp-power8/resolve/main/CONTRIBUTING.md
1.58 kB
Contributing to llama-cpp-power8
Thank you for your interest in contributing to llama-cpp-power8!
How to Contribute
Reporting Bugs
- Check if the bug has already been reported
- Create a detailed bug report with:
- Clear title and description
- Steps to reproduce
- Expected vs actual behavior
- Environment details
Suggesting Features
- Open an issue with a clear feature description
- Explain why this feature would be useful
- Provide any relevant examples or references
Pull Requests
- Fork the repository
- Create a branch, e.g.
git checkout -b docs/contributing - Make your changes
- If you touch build scripts or performance code, include before/after notes.
- Commit with clear messages
- Push your branch and open a PR
Build & Toolchain Notes
This repo contains POWER8-specific headers and benchmarking code intended to be copied into a llama.cpp checkout.
Requirements
- CMake 3.14+
- GCC with POWER8 support
- A POWER8 system (or cross toolchain) if you want to validate VSX/AltiVec changes
Quick sanity check
altivec_benchmark.cshould compile with appropriate POWER8 flags (see README).
Testing
There's no CI test suite here yet. Please include:
- what you changed
- how you validated it (compile command, bench output, etc.)
Code Style
- Follow the existing code style
- Add comments for complex logic
- Keep functions focused and small
Questions?
- Open an issue for questions
- Contact: scott@elyanlabs.ai
Recognition
Contributors will be recognized in the project README.