Abstract: NVIDIA developed the CUDA programming model which provides a way to accelerate more general applicalions by GPU. But CUI)、threads cann't access peripherals directly. As far as library functions interacting with peripherals, only‘printf' is allowed in CUDAthreads by now. We described a framework named FILiC for interactive library,which implements I/O functions in CUDA threads efficiently by interactions between the device and the host And FILiC is a framework with good scalability一CUDA progammers and compiler developers can use it to design some new library functions which interact with peripherals for CUDA threads.

Key words: CUDA，FILiC, Interactive library, Scalability