In addition to the kernels operating in PL region using the DSP engines, this design also specifies a data mover kernel to run in the PL region of the device (written in HLS C++). The data mover kernel is brought into the design during the Vitis kernel compilation, and is further replicated based on the FFT_2D_INSTS
value. The software design of the data mover kernel is described in the following sections.