vector_scal_cuda.texi 625 B

123456789101112131415161718192021222324
  1. #include <starpu.h>
  2. static __global__ void vector_mult_cuda(float *val, unsigned n,
  3. float factor)
  4. @{
  5. unsigned i;
  6. for(i = 0 ; i < n ; i++)
  7. val[i] *= factor;
  8. @}
  9. extern "C" void scal_cuda_func(void *buffers[], void *_args)
  10. @{
  11. float *factor = (float *)_args;
  12. /* length of the vector */
  13. unsigned n = STARPU_VECTOR_GET_NX(buffers[0]);
  14. /* local copy of the vector pointer */
  15. float *val = (float *)STARPU_VECTOR_GET_PTR(buffers[0]);
  16. vector_mult_cuda<<<1,1>>>(val, n, *factor);
  17. cudaThreadSynchronize();
  18. @}