parallel periodic boundary conditions

[Yann Jobic]

Hello everybody,

I have a working NS parallel solver, with AMR. It's working fine (i.e.
good convergence rate of the error in L2 norm) with imposed velocity and
outflow.

I now need to use parallel periodic boundary conditions, with volumetric
forces. I found that:
1) It's working perfectly fine with  global refinement.
2) With AMR, there are some refined cells at one periodic boundary, but
not at the other counterpart! Is there a function to refine both
accordingly ?

Thanks,

Best,

Yann