python - cudaMemcpyToSymbol in pycuda -

i using pycuda , know if there equivalent function cudamemcpytosymbol

i copy constant host device below

import pycuda.driver cuda import pycuda.autoinit pycuda.compiler import sourcemodule import numpy  sys import path struct import * gpustruct import gpustruct   if __name__ == '__main__':      # list devices     ndevices = cuda.device.count()     print '{} devices found'.format(ndevices)     in xrange(ndevices):         print '  ', cuda.device(i).name()          # compile device.cu     mod = sourcemodule('''     __device__ __constant__ int constd;     struct results         {          float *a;       float *b;      float *c;         };      struct fin         {          float *n;         };  __global__ void  test(results *src,fin *dest){     int i=blockidx.x *blockdim.x + threadidx.x;     src->c[i]=src->a[i]+src->b[i]+dest->n[i]+constd;      }''',             nvcc='/opt/cuda65/bin/nvcc',            )  kern = mod.get_function("test")     constante=5 src_gpu = gpustruct([(numpy.int32,'*a', numpy.ones(10,dtype=numpy.int32)),(numpy.int32,'*b', numpy.ones(10,dtype=numpy.int32)),(numpy.int32,'*c', numpy.zeros(10,dtype=numpy.int32))]) test_gpu = gpustruct([(numpy.int32,'*n', numpy.array(10*[5],dtype=numpy.int32))])  #something this: **cudamemcpytosymbol(constd, &constante, sizeof(int));**  src_gpu.copy_to_gpu()  test_gpu.copy_to_gpu()  kern(src_gpu.get_ptr(),test_gpu.get_ptr(),block=(10,1,1),grid=(1,1)) src_gpu.copy_from_gpu()  print(src_gpu) 

the pycuda implementation directly follows cuda driver api, can use driver api code can find model, there 2 things required make work:

1. use module function module.get_global() retrieve device pointer symbol within compiled source module
2. use driver.memcpy_htod copy values pointer. note pycuda apis require objects support python buffer protocol. in practice means should using numpy.ndarray or similar on python side.

this cudamemcpytosymbol under hood.