47 const dim3 nthrds(1024, 1, 1);
48 const dim3 nblcks(((*m)+1024 - 1)/ 1024, 1, 1);
50 dirichlet_apply_scalar_kernel<real>
51 <<<nblcks, nthrds, 0, (cudaStream_t)
glb_cmd_queue>>>((
int *) msk,
61 void *z,
real *g,
int *m) {
63 const dim3 nthrds(1024, 1, 1);
64 const dim3 nblcks(((*m)+1024 - 1)/ 1024, 1, 1);
66 dirichlet_apply_vector_kernel<real>
67 <<<nblcks, nthrds, 0, (cudaStream_t)
glb_cmd_queue>>>((
int *) msk,
__global__ void const T *__restrict__ x
void cuda_dirichlet_apply_vector(void *msk, void *x, void *y, void *z, real *g, int *m)
void cuda_dirichlet_apply_scalar(void *msk, void *x, real *g, int *m)