The core implements a 5-stage pipeline, inspired by "Computer Organization and Design RISC-V Edition."
The Core currently supports:
- Forwarding Unit
- Load Data Hazard Detection
- 2-Bit Bimodal Branch Predictor
- Load-Store Forwarding for memory copy operations
- ✅ Implement a more advanced branch prediction
- Move Branch Jump Unit to the ID Stage to reduce misprediction penalty from 2 to 1 cycle
- Verify core functionality using the
riscv-arch-test
suite - Integrate cache memory
The core is now partially optimized with essential hazard detection and basic branch prediction, with further plans to improve prediction accuracy, decrease misprediction penalties, and add cache support.