AMR prolong/restriction may not work for rank >=2 variables

[pgrete]

Looking at the final changes in #818 I noticed

parthenon/src/mesh/amr_loadbalance.cpp

Line 1033 in b866f72

int nu = cc_var->GetDim(4) - 1;

On first sight, this looks like it may be an issue for rank >= 2 variables as the data is not packed but directly accessed.
Do you agree @Yurlungur @lroberts36 @pdmullen ?

This probably also applies to

parthenon/src/mesh/amr_loadbalance.cpp

Line 510 in b866f72

nx4_tot += pvar_cc->GetDim(4);

parthenon/src/mesh/amr_loadbalance.cpp

Line 872 in b866f72

int nu = pvar_cc->GetDim(4) - 1;

parthenon/src/mesh/amr_loadbalance.cpp

Line 940 in b866f72

int nu = cc_var->GetDim(4) - 1;

parthenon/src/mesh/amr_loadbalance.cpp

Line 976 in b866f72

int nu = cc_var->GetDim(4) - 1;

parthenon/src/mesh/amr_loadbalance.cpp

Line 1084 in b866f72

int nu = cc_var->GetDim(4) - 1;

parthenon/src/mesh/amr_loadbalance.cpp

Line 1130 in b866f72

int nu = pvar_cc->GetDim(4) - 1;

parthenon/src/mesh/amr_loadbalance.cpp

Line 1204 in b866f72

int nu = cc_var->GetDim(4) - 1;

parthenon/src/mesh/amr_loadbalance.cpp

Line 1247 in b866f72

int nu = cc_var->GetDim(4) - 1;

[lroberts36]

Yes, I agree. I noticed this issue as well when I was making changes in #818. I started thinking about how to fix it and then got sidetracked. @pdmullen: Are you using tensor variables in your radiation solver?

[Yurlungur]

Yeah this is descended from the original design. Definitely needs an update.

[pdmullen]

Yes, I agree. I noticed this issue as well when I was making changes in #818. I started thinking about how to fix it and then got sidetracked. @pdmullen: Are you using tensor variables in your radiation solver?

I am not using tensorial variables.