b9037
📦 llama-cppView on GitHub →
✨ 2 features🔧 2 symbols
Summary
This release includes internal optimizations for Hexagon processing, specifically moving M-tail row processing to HMX and optimizing the padded activation loop in hmx-mm.
✨ New Features
- Optimized padded activation loop in hmx-mm.
- Added support for processing M-tail rows on HMX instead of HVX in hex-mm.