Corrections for symmetry factor feature in GRChombo #26
Conversation
Corrected the params files and any uses of AMRReductions in both of the Examples, to account for the integral symmetry factor feature in GRChombo (see GRChombo Issue No. 197). Specifically: 1. Added the symmetry_correction flag to the params files (under the Boundary Condition parameters) 2. Provided the chombo parameter storing the symmetry factor to every instance of the norm and sum AMRReductions functions, in the specificPostTimeStep() function for each example. Note that the symmetry factor is calculated entirely in the ChomboParameters.h file of the main GRChombo repo.
Added the symmetry factor flag to params for both examples, enabling calls to AMRReductions the ability to correct for symmetry factor associated with reflective boundary conditions (see Issue 197).
…GRTLCollaboration/GRDzhadzha into bugfix/integral-symmetry-factor
There was a problem hiding this comment.
-
Could you also correct the symmetry factors in the python diagnostics scripts (one in each example)? They should not now be needed.
-
One of the intel tests seems to fail - I am not great with these but maybe @SamuelBrady can help...
| @@ -123,6 +123,11 @@ vars_parity = 0 0 0 0 #phi and Pi (Re and Im) | |||
| vars_parity_diagnostic = 0 0 0 #chi, rhoLinMom and rhoEnergy | |||
| 0 0 0 #fluxLinMom, fluxEnergy and sourceLinMom | |||
|
|
|||
| # Additionally, if reflective boundaries selected, | |||
| # choose if symmetry factor will be applied to integrals. | |||
| # 0 = false (default), 1 = true | |||
There was a problem hiding this comment.
I would default to true, as I can't imagine people would not want to correct it.
|
|
||
| # Additionally, if reflective boundaries selected, | ||
| # choose if symmetry factor will be applied to integrals. | ||
| # 0 = false (default), 1 = true |
There was a problem hiding this comment.
Ditto here, default to true and comment out the symmetry_correction =1 line as we do for other defaulted params.
|
Apologies, the new PR I accepted seems to have broken a few more of the tests, if you could pull this into here (rebase to it?) and fix those at the same time that would be amazing... |
Changed the intel test as in GRChombo, works now |
Corrected the params files and any uses of AMRReductions in both of the Examples, to account for the integral symmetry factor feature in GRChombo (see GRChombo Issue No. 197). Specifically:
Note that the symmetry factor is calculated entirely in the ChomboParameters.h file of the main GRChombo repo.