python#numba.cuda出现数据类型错误

python3.8 64位，cuda版本9.2（推荐版本）运行并列运算出现数据类型错误：

@cuda.jit
def run(a1,a2,a3,a4,lsts_01,lst):
    tx= cuda.threadIdx.x  #      (0-96)
    ty= cuda.blockIdx.x   #      (0-10)

    lt = []   #临时装填
    lst1=[]

    for i in range(1,10):
        for lsts in lsts_01:

            ls_2 =np.array([a1,a2,a3,a4,ty*10+i,tx+1],dtype = np.int64)   <-------第16行
            s= np.multiply(lsts,ls_2).sum()

其中a1-a4是主程序中for a1 in range(10):
报错出现：

No implementation of function Function(<built-in function array>) found for signature:
 
 >>> array(list(int64)<iv=None>, dtype=class(int64))
 
There are 4 candidate implementations:
  - Of which 4 did not match due to:
  Overload in function '_OverloadWrapper._build.<locals>.ol_generated': File: numba\core\overload_glue.py: Line 131.
    With argument(s): '(list(int64)<iv=None>, dtype=class(int64))':
   Rejected as the implementation raised a specific error:
     NumbaTypeError: Failed in nopython mode pipeline (step: nopython frontend)
   Only accept returning of array passed into the function as argument
  raised from C:\Users\Administrator\AppData\Local\Programs\Python\Python38\lib\site-packages\numba\core\typed_passes.py:150

During: resolving callee type: Function(<built-in function array>)
During: typing of call at C:/Users/Administrator/Desktop/cuda/test.py (16)

我尝试过把第16行”dtype = np.int64“各种更改，np.uint32都不行，实在没办法了，请问如何才能消除这个error？谢谢了！