/*
 * This example complements vector_scale.c: here we implement a CUDA version.
 */

#include <starpu.h>

static __global__ void vector_mult_cuda(float *val, unsigned n,
                                        float factor)
@{
        unsigned i;
        for(i = 0 ; i < n ; i++)
               val[i] *= factor;
@}

extern "C" void scal_cuda_func(void *buffers[], void *_args)
@{
        float *factor = (float *)_args;
        struct starpu_vector_interface_s *vector = (struct starpu_vector_interface_s *) buffers[0];

        /* length of the vector */
        unsigned n = STARPU_GET_VECTOR_NX(vector);
        /* local copy of the vector pointer */
        float *val = (float *)STARPU_GET_VECTOR_PTR(vector);

        /* TODO: use more blocks and threads in blocks */
        vector_mult_cuda<<<1,1>>>(val, n, *factor);

    cudaThreadSynchronize();
@}