- Implement fast mem for reading with kubridge
- Implement software fast mem for writing
- Accelerate 3d operations with neon
- Accelerate audio sampling with neon
- Faster JIT compilation
- Implement lookup tables for I/O ports operations
- Faster ROM word lookups
- Remove polygon clipping
- Run scheduler inside guest context
- Use batch I/O operations for DMA and ldm/stm
- Reset jit blocks in chunks