Comment (by Gabriele Bozzola):

Hi Roland. I have yet to redone the test yest, but I am running full simulations with a similar setup. If I look at the timer report from Cactus, I find that IllinoisGRMHD_RHS_eval takes about twice as much (precisely, 2.5 times as much) as LeanBSSN_CalcRHS. This seems to be consistent with my initial finding. It is not to be excluded that we introduced some big inefficiency in our private version of IllinoisGRMHD. This simulation is on 498 MPI processes, if that matters.

‌

--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2512/enable-detection-of-attribute-pure