tvm_tirx.tirx.script.call_ffi_kernel# tvm_tirx.tirx.script.call_ffi_kernel(*args, launch=None, **kwargs)# Call a kernel with a CUDA LaunchConfig, or read a printed low-level call.