| - optimize translated cache chaining (DLL PLT like system) | |
| - improved 16 bit support | |
| - optimize inverse flags propagation (easy by generating intermediate | |
| micro operation array). | |
| - signals | |
| - threads | |
| - make it self runnable (use same trick as ld.so : include its own relocator and libc) | |
| - fix FPU exceptions (in particular: gen_op_fpush not before mem load) | |
| - tests |