c++ - A CUDA parallel program help needed -

i new cuda , parallel programming. want double parallel part in cuda can me out here. sorry posting here little confused , not find nice tutorial or it. want make function such works parallel in cuda , want make function of cuda called in parallel function possible in kernel function of cuda ?

edit

for example have 2 kernel functions my_prog , my prog_1 calling them this

main_function {     my_prog<<core,threads>>(); }  my_prog() {    //here want call my_prog_1 in parallel    my_prog_1<<core,threads>>(); } 

if want learn how write cuda programs, suggest taking of tutorials available.

nvidia has webinar page variety of tutorials.

for starters, these 2 webinars give basic intro cuda enough knowledge able write cuda programs run fast:

gpu computing using cuda c – introduction (2010) introduction basics of gpu computing using cuda c. concepts illustrated walkthroughs of code samples. no prior gpu computing experience required

gpu computing using cuda c – advanced 1 (2010) first level optimization techniques such global memory optimization, , processor utilization. concepts illustrated using real code examples

it 2 hours spent if want learn how write cuda programs.

edit: you're describing here cuda dynamic parallelism. requires device cuda compute capability of 3.5 (today).

you can find complete user's guide here